chore: release v2.57.0

* infra: Enable Porter Application worker (#2434)

* Create porter_stack_web.yml file

* push

* push

---------

Co-authored-by: porter-deployment-app[bot] <87230664+porter-deployment-app[bot]@users.noreply.github.com>
Co-authored-by: Max Deichmann <m.deichmann@tum.de>

.devcontainer/devcontainer.json

@@ -0,0 +1,14 @@
+{
+	"name": "langfuse dev",
+    "image": "mcr.microsoft.com/devcontainers/javascript-node:20-bookworm",
+    "features": {
+        "ghcr.io/devcontainers/features/docker-in-docker:2": {
+        "version": "latest",
+        "moby": true
+        }
+    },
+    "forwardPorts": [3000],
+
+    "onCreateCommand": "npm install -g pnpm@8.15.5",
+    "postCreateCommand": "cp .env.dev.example .env && pnpm run dx-f"
+}

.env.dev.example

@@ -34,10 +34,6 @@ S3_SECRET_ACCESS_KEY=
 S3_BUCKET_NAME=
 S3_REGION=
 
-# Prompt playground
-OPENAI_API_KEY=""
-ANTHROPIC_API_KEY=""
-
 # Set during docker build of application
 # Used to disable environment verification at build time
 # DOCKER_BUILD=1
@@ -47,5 +43,5 @@ REDIS_PORT=6379
 REDIS_AUTH="myredissecret"
 
 LANGFUSE_WORKER_PASSWORD=mybasicauthsecret
-# openssl rand -base64 32 used only here
+# openssl rand -hex 32 used only here
 ENCRYPTION_KEY=6c16874e5c0f0cc74ddec00425fa99fbe9ffbe412b7d5d906a4b00005df91403 

.env.local.example

@@ -13,11 +13,6 @@ NEXTAUTH_URL="http://localhost:3000"
 LANGFUSE_ENABLE_EXPERIMENTAL_FEATURES="false"
 SALT="salt"
 
-
-# Prompt playground
-OPENAI_API_KEY=""
-ANTHROPIC_API_KEY=""
-
 # Redis
 REDIS_HOST="127.0.0.1"
 REDIS_PORT=6379

.env.prod.example

@@ -43,6 +43,7 @@ LANGFUSE_CSP_ENFORCE_HTTPS="true"
 # AUTH_GOOGLE_CLIENT_ID=
 # AUTH_GOOGLE_CLIENT_SECRET=
 # AUTH_GOOGLE_ALLOW_ACCOUNT_LINKING=false
+# AUTH_GOOGLE_ALLOWED_DOMAINS=langfuse.com,google.com # optional allowlist of workspace domains that can sign in via Google
 # AUTH_GITHUB_CLIENT_ID=
 # AUTH_GITHUB_CLIENT_SECRET=
 # AUTH_GITHUB_ALLOW_ACCOUNT_LINKING=false
@@ -58,6 +59,15 @@ LANGFUSE_CSP_ENFORCE_HTTPS="true"
 # AUTH_AUTH0_CLIENT_SECRET=
 # AUTH_AUTH0_ISSUER=
 # AUTH_AUTH0_ALLOW_ACCOUNT_LINKING=false
+# AUTH_COGNITO_CLIENT_ID=
+# AUTH_COGNITO_CLIENT_SECRET=
+# AUTH_COGNITO_ISSUER=
+# AUTH_COGNITO_ALLOW_ACCOUNT_LINKING=false
+# AUTH_CUSTOM_CLIENT_ID=
+# AUTH_CUSTOM_CLIENT_SECRET=
+# AUTH_CUSTOM_ISSUER=
+# AUTH_CUSTOM_NAME=
+# AUTH_CUSTOM_ALLOW_ACCOUNT_LINKING=false
 
 # Transactional email, optional
 # Defines the email address to use as the from address.
@@ -71,6 +81,7 @@ LANGFUSE_CSP_ENFORCE_HTTPS="true"
 # S3_SECRET_ACCESS_KEY=
 # S3_BUCKET_NAME=
 # S3_REGION=
+# BATCH_EXPORT_DOWNLOAD_LINK_EXPIRATION_HOURS=
 
 # Exports are streamed to S3 in pages to avoid memory issues
 # The page size can be adjusted if needed to optimize performance
@@ -118,10 +129,6 @@ LANGFUSE_CSP_ENFORCE_HTTPS="true"
 # NEXT_PUBLIC_TURNSTILE_SITE_KEY=
 # TURNSTILE_SECRET_KEY=
 
-# Prompt playground
-# OPENAI_API_KEY=""
-# ANTHROPIC_API_KEY=""
-
 # Betterstack
 # LANGFUSE_TEAM_BETTERSTACK_TOKEN=
 

.github/workflows/pipeline.yml

@@ -40,25 +40,34 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v3
 
-      - name: Build Web Docker image
-        uses: docker/build-push-action@v4
-        with:
-          context: .
-          file: ./web/Dockerfile
-          push: false
+      - name: Build and run both images from compose
+        run: |
+          docker compose -f docker-compose.build.yml up -d
+          sleep 5 # Wait for PostgreSQL to accept connections
 
-      - name: Build Worker Docker image
-        uses: docker/build-push-action@v4
-        with:
-          context: .
-          file: ./worker/Dockerfile
-          push: false
+      - name: Check server health
+        run: |
+          timeout 10 bash -c 'until curl -f http://localhost:3000/api/public/health; do sleep 2; done'
+
+      - name: Check worker health
+        run: |
+          timeout 10 bash -c 'until curl -f http://localhost:3030/api/health; do sleep 2; done'
+
+      - name: Ensure no unhealthy status
+        run: |
+          if docker-compose ps | grep "(unhealthy)"; then
+            echo "One or more services are unhealthy"
+            exit 1
+          else
+            echo "All services are healthy"
+          fi
 
   tests-web:
     runs-on: ubuntu-latest
     strategy:
       matrix:
         node-version: [20]
+        postgres-version: [12, 15]
     steps:
       - name: Set Swap Space
         uses: pierotofy/set-swap-space@master
@@ -89,6 +98,9 @@ jobs:
         run: |
           docker compose -f docker-compose.dev.yml up -d
           sleep 5 # Wait for PostgreSQL to accept connections
+          docker compose ps
+        env:
+          POSTGRES_VERSION: ${{ matrix.postgres-version }}
 
       - name: Seed DB
         run: |
@@ -109,6 +121,7 @@ jobs:
     strategy:
       matrix:
         node-version: [20]
+        postgres-version: [12, 15]
     steps:
       - name: Set Swap Space
         uses: pierotofy/set-swap-space@master
@@ -139,6 +152,7 @@ jobs:
         run: |
           docker compose -f docker-compose.dev.yml up -d
           sleep 5 # Wait for PostgreSQL to accept connections
+          docker compose ps
 
       - name: Seed DB
         run: |
@@ -176,6 +190,7 @@ jobs:
       - name: Run + migrate
         run: |
           docker compose -f docker-compose.dev.yml up -d
+          docker compose ps
           sleep 5 # Wait for PostgreSQL to accept connections
 
       - name: Seed DB

.github/workflows/porter_stack_web_staging.yml

@@ -0,0 +1,29 @@
+"on":
+  push:
+    branches:
+      - main
+name: Deploy to web (Staging)
+jobs:
+  porter-deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+      - name: Set Github tag
+        id: vars
+        run: echo "sha_short=$(git rev-parse --short HEAD)" >> $GITHUB_OUTPUT
+      - name: Setup porter
+        uses: porter-dev/setup-porter@v0.1.0
+      - name: Deploy stack
+        timeout-minutes: 30
+        run: exec porter apply
+        env:
+          PORTER_CLUSTER: "4037"
+          PORTER_DEPLOYMENT_TARGET_ID: c22d8fa2-f97c-4e1f-9cae-83cc6bfbad35
+          PORTER_HOST: https://dashboard.getporter.dev
+          PORTER_PR_NUMBER: ${{ github.event.number }}
+          PORTER_PROJECT: "12565"
+          PORTER_REPO_NAME: ${{ github.event.repository.name }}
+          PORTER_STACK_NAME: web
+          PORTER_TAG: ${{ steps.vars.outputs.sha_short }}
+          PORTER_TOKEN: ${{ secrets.PORTER_STACK_12565_4037 }}

.github/workflows/porter_stack_web_us_temp.yml

@@ -0,0 +1,29 @@
+"on":
+  push:
+    branches:
+      - production
+name: Deploy to web (US temp)
+jobs:
+  porter-deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+      - name: Set Github tag
+        id: vars
+        run: echo "sha_short=$(git rev-parse --short HEAD)" >> $GITHUB_OUTPUT
+      - name: Setup porter
+        uses: porter-dev/setup-porter@v0.1.0
+      - name: Deploy stack
+        timeout-minutes: 30
+        run: exec porter apply
+        env:
+          PORTER_CLUSTER: "4054"
+          PORTER_DEPLOYMENT_TARGET_ID: cddc0f2a-bb19-401a-9819-29f59f76b101
+          PORTER_HOST: https://dashboard.porter.run
+          PORTER_PR_NUMBER: ${{ github.event.number }}
+          PORTER_PROJECT: "12565"
+          PORTER_REPO_NAME: ${{ github.event.repository.name }}
+          PORTER_STACK_NAME: web
+          PORTER_TAG: ${{ steps.vars.outputs.sha_short }}
+          PORTER_TOKEN: ${{ secrets.PORTER_STACK_12565_4054 }}

.github/workflows/porter_stack_worker_staging.yml

@@ -0,0 +1,29 @@
+"on":
+  push:
+    branches:
+      - main
+name: Deploy to worker (staging)
+jobs:
+  porter-deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+      - name: Set Github tag
+        id: vars
+        run: echo "sha_short=$(git rev-parse --short HEAD)" >> $GITHUB_OUTPUT
+      - name: Setup porter
+        uses: porter-dev/setup-porter@v0.1.0
+      - name: Deploy stack
+        timeout-minutes: 30
+        run: exec porter apply
+        env:
+          PORTER_CLUSTER: "4037"
+          PORTER_DEPLOYMENT_TARGET_ID: c22d8fa2-f97c-4e1f-9cae-83cc6bfbad35
+          PORTER_HOST: https://dashboard.getporter.dev
+          PORTER_PR_NUMBER: ${{ github.event.number }}
+          PORTER_PROJECT: "12565"
+          PORTER_REPO_NAME: ${{ github.event.repository.name }}
+          PORTER_STACK_NAME: worker
+          PORTER_TAG: ${{ steps.vars.outputs.sha_short }}
+          PORTER_TOKEN: ${{ secrets.PORTER_STACK_12565_4037 }}

.github/workflows/porter_stack_worker_us.yml

@@ -0,0 +1,29 @@
+"on":
+  push:
+    branches:
+      - production
+name: Deploy to worker (US)
+jobs:
+  porter-deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+      - name: Set Github tag
+        id: vars
+        run: echo "sha_short=$(git rev-parse --short HEAD)" >> $GITHUB_OUTPUT
+      - name: Setup porter
+        uses: porter-dev/setup-porter@v0.1.0
+      - name: Deploy stack
+        timeout-minutes: 30
+        run: exec porter apply
+        env:
+          PORTER_CLUSTER: "4038"
+          PORTER_DEPLOYMENT_TARGET_ID: 260cd04e-5dc8-4b22-a291-ad4453f981f7
+          PORTER_HOST: https://dashboard.getporter.dev
+          PORTER_PR_NUMBER: ${{ github.event.number }}
+          PORTER_PROJECT: "12565"
+          PORTER_REPO_NAME: ${{ github.event.repository.name }}
+          PORTER_STACK_NAME: worker
+          PORTER_TAG: ${{ steps.vars.outputs.sha_short }}
+          PORTER_TOKEN: ${{ secrets.PORTER_STACK_12565_4038 }}

.github/workflows/porter_stack_worker_us_temp.yml

@@ -0,0 +1,29 @@
+"on":
+  push:
+    branches:
+      - production
+name: Deploy to worker (US temp)
+jobs:
+  porter-deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+      - name: Set Github tag
+        id: vars
+        run: echo "sha_short=$(git rev-parse --short HEAD)" >> $GITHUB_OUTPUT
+      - name: Setup porter
+        uses: porter-dev/setup-porter@v0.1.0
+      - name: Deploy stack
+        timeout-minutes: 30
+        run: exec porter apply
+        env:
+          PORTER_CLUSTER: "4054"
+          PORTER_DEPLOYMENT_TARGET_ID: cddc0f2a-bb19-401a-9819-29f59f76b101
+          PORTER_HOST: https://dashboard.porter.run
+          PORTER_PR_NUMBER: ${{ github.event.number }}
+          PORTER_PROJECT: "12565"
+          PORTER_REPO_NAME: ${{ github.event.repository.name }}
+          PORTER_STACK_NAME: worker
+          PORTER_TAG: ${{ steps.vars.outputs.sha_short }}
+          PORTER_TOKEN: ${{ secrets.PORTER_STACK_12565_4054 }}

.github/workflows/snyk.yml

@@ -0,0 +1,39 @@
+name: Snyk Container
+on: push
+jobs:
+  snyk:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - name: Build a Docker image
+        run: docker compose -f docker-compose.build.yml up -d
+
+      - name: Run Snyk to check Docker image for vulnerabilities (langfuse-server)
+        continue-on-error: true
+        uses: snyk/actions/docker@master
+        env:
+          SNYK_TOKEN: ${{ secrets.SNYK_TOKEN }}
+        with:
+          image: langfuse-server
+          args: --file=web/Dockerfile
+
+      - name: Upload result to GitHub Code Scanning
+        uses: github/codeql-action/upload-sarif@v2
+        with:
+          sarif_file: snyk.sarif
+          category: web
+
+      - name: Run Snyk to check Docker image for vulnerabilities (langfuse-worker)
+        continue-on-error: true
+        uses: snyk/actions/docker@master
+        env:
+          SNYK_TOKEN: ${{ secrets.SNYK_TOKEN }}
+        with:
+          image: langfuse-worker
+          args: --file=worker/Dockerfile
+
+      - name: Upload result to GitHub Code Scanning
+        uses: github/codeql-action/upload-sarif@v2
+        with:
+          sarif_file: snyk.sarif
+          category: worker

.husky/pre-commit

@@ -0,0 +1,22 @@
+#!/bin/sh
+
+# Get the current branch
+current_branch=$(git rev-parse --abbrev-ref HEAD)
+
+# Define the protected branch
+protected_branch="main"
+
+# Check if the current branch is the protected branch
+if [ "$current_branch" = "$protected_branch" ]; then
+    echo "🚨 You are about to commit to the $protected_branch branch. Are you sure? (y/n)"
+    read -r answer < /dev/tty
+    if [ "$answer" != "${answer#[Yy]}" ]; then
+        exit 0 # Commit will proceed
+    else
+        echo "Commit to $protected_branch branch has been canceled."
+        exit 1 # Commit will be blocked
+    fi
+fi
+
+# If not the protected branch, proceed with the commit
+exit 0

.husky/pre-push

@@ -0,0 +1,22 @@
+#!/bin/sh
+
+# Get the current branch
+current_branch=$(git rev-parse --abbrev-ref HEAD)
+
+# Define the protected branch
+protected_branch="main"
+
+# Check if the current branch is the protected branch
+if [ "$current_branch" = "$protected_branch" ]; then
+    echo "🚨 You are about to push to the $protected_branch branch. Are you sure? (y/n)"
+    read -r answer < /dev/tty
+    if [ "$answer" != "${answer#[Yy]}" ]; then
+        exit 0 # Push will proceed
+    else
+        echo "Push to $protected_branch branch has been canceled."
+        exit 1 # Push will be blocked
+    fi
+fi
+
+# If not the protected branch, proceed with the push
+exit 0

CONTRIBUTING.md

@@ -55,7 +55,7 @@ A good first step is to search for open [issues](https://github.com/langfuse/lan
 
 > [!NOTE]
 > Infrastructure will change in Langfuse version 3.0. More in the [GitHub Discussions](https://github.com/orgs/langfuse/discussions/1902).
-> `langfuse/langfuse/worker` is under active development and not recommended for production use in Langfuse 2.x
+> `langfuse/langfuse/worker` is under active development and not recommended for production use in Langfuse 2.x.
 
 ```mermaid
 flowchart TB
@@ -151,6 +151,8 @@ Requirements
 - Node.js 20 as specified in the [.nvmrc](.nvmrc)
 - Docker to run the database locally
 
+**Note:** You can also simply run Langfuse in a **GitHub Codespace** via the provided devcontainer. To do this, click on the green "Code" button in the top right corner of the repository and select "Open with Codespaces".
+
 **Steps**
 
 1. Fork the repository and clone it locally
@@ -191,6 +193,16 @@ Requirements
     pnpm run dev
    ```
 
+7. Open the web app in the browser:
+
+   http://localhost:3000
+
+8. Log in as a test user (after you ran `db:seed` command):
+
+   Username: demo@langfuse.com
+
+   Password: password
+
 ## Monorepo quickstart
 
 - Available packages and their dependencies
@@ -358,7 +370,9 @@ The background color of the following component will be `hsl(var(--primary))` an
 
 By following these guidelines, you can ensure that any contributions to our theme are consistent, maintainable, and aligned with our design system.
 
-## Using secrets stored in 1Password
+## Maintainers
+
+### Using secrets stored in 1Password
 
 When applying changes to non-local environments, you may need to use secrets stored in 1Password. We use the 1Password CLI for this purpose.
 

README.md

@@ -24,6 +24,8 @@
    </div>
    <span>Langfuse uses <a href="https://github.com/orgs/langfuse/discussions"><strong>Github Discussions</strong></a>  for Support and Feature Requests.</span>
    <br/>
+   <span>We're hiring. <a href="https://langfuse.com/careers"><strong>Join us</strong></a> in Backend Engineering, Product Engineering, and Developer Relations.</span>
+   <br/>
    <br/>
    <div>
       <a href="https://github.com/langfuse/langfuse/blob/main/LICENSE"><img src="https://img.shields.io/badge/License-MIT-red.svg?style=flat-square" alt="MIT License"></a>
@@ -106,22 +108,26 @@ See the [→ Quickstart](https://langfuse.com/docs/get-started) to integrate Lan
 
 ### Integrations
 
-| Integration                                                      | Supports      | Description                                                 |
-| ---------------------------------------------------------------- | ------------- | ----------------------------------------------------------- |
-| [**SDK** - _recommended_](https://langfuse.com/docs/sdk)         | Python, JS/TS | Manual instrumentation using the SDKs for full flexibility. |
-| [OpenAI SDK](https://langfuse.com/docs/integrations/openai)      | Python, JS/TS | Automated instrumentation of OpenAI SDK.                    |
-| [Langchain](https://langfuse.com/docs/integrations/langchain)    | Python, JS/TS | Instrumentation via Langchain callbacks.                    |
-| [LlamaIndex](https://langfuse.com/docs/integrations/llama-index) | Python        | Automated instrumentation via LlamaIndex callback system.   |
-| [API](https://langfuse.com/docs/api)                             |               | Directly call the public API. OpenAPI spec available.       |
+| Integration                                              | Supports                   | Description                                                                                                                                      |
+| -------------------------------------------------------- | -------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ |
+| [SDK](/docs/sdk)                                         | Python, JS/TS              | Manual instrumentation using the SDKs for full flexibility.                                                                                      |
+| [OpenAI](/docs/integrations/openai)                      | Python, JS/TS              | Automated instrumentation using drop-in replacement of OpenAI SDK.                                                                               |
+| [Langchain](/docs/integrations/langchain)                | Python, JS/TS              | Automated instrumentation by passing callback handler to Langchain application.                                                                  |
+| [LlamaIndex](/docs/integrations/llama-index/get-started) | Python                     | Automated instrumentation via LlamaIndex callback system.                                                                                        |
+| [Haystack](/docs/integrations/haystack)                  | Python                     | Automated instrumentation via Haystack content tracing system.                                                                                   |
+| [LiteLLM](/docs/integrations/litellm)                    | Python, JS/TS (proxy only) | Use any LLM as a drop in replacement for GPT. Use Azure, OpenAI, Cohere, Anthropic, Ollama, VLLM, Sagemaker, HuggingFace, Replicate (100+ LLMs). |
+| [API](/docs/api)                                         |                            | Directly call the public API. OpenAPI spec available.                                                                                            |
 
-External projects/packages that integrate with Langfuse:
+Packages that integrate with Langfuse:
 
-| Name                                                            | Description                                                                                                                                      |
-| --------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ |
-| [LiteLLM](https://langfuse.com/docs/integrations/litellm)       | Use any LLM as a drop in replacement for GPT. Use Azure, OpenAI, Cohere, Anthropic, Ollama, VLLM, Sagemaker, HuggingFace, Replicate (100+ LLMs). |
-| [Flowise](https://langfuse.com/docs/integrations/flowise)       | JS/TS no-code builder for customized LLM flows.                                                                                                  |
-| [Langflow](https://langfuse.com/docs/integrations/langflow)     | Python-based UI for LangChain, designed with react-flow to provide an effortless way to experiment and prototype flows.                          |
-| [Superagent](https://langfuse.com/docs/integrations/superagent) | Open Source AI Assistant Framework & API for prototyping and deployment of agents.                                                               |
+| Name                                                       | Description                                                                                                             |
+| ---------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------- |
+| [Instructor](/docs/integrations/instructor)                | Library to get structured LLM outputs (JSON, Pydantic)                                                                  |
+| [Mirascope](/docs/integrations/mirascope)                  | Python toolkit for building LLM applications.                                                                           |
+| [AI SDK by Vercel](/docs/sdk/typescript/example-vercel-ai) | Typescript SDK that makes streaming LLM outputs super easy.                                                             |
+| [Flowise](/docs/integrations/flowise)                      | JS/TS no-code builder for customized LLM flows.                                                                         |
+| [Langflow](/docs/integrations/langflow)                    | Python-based UI for LangChain, designed with react-flow to provide an effortless way to experiment and prototype flows. |
+| [Superagent](/docs/integrations/superagent)                | Open Source AI Assistant Framework & API for prototyping and deployment of agents.                                      |
 
 ## Questions and feedback
 

careers

@@ -3,9 +3,9 @@ Join us in scaling Langfuse in Berlin, Germany. We are an open source company, w
 
 _Open Roles_
 
-- Backend Engineer, 70-110k EUR, 0.25-0.75% Equity, https://www.ycombinator.com/companies/langfuse/jobs/mnrdwla-backend-engineer
-- Product Engineer, 70-110k EUR, 0.25-0.75% Equity, https://www.ycombinator.com/companies/langfuse/jobs/aAvmoFB-product-engineer
-- Developer Advocate, 60-100k EUR, 0.25-0.5% Equity, https://www.ycombinator.com/companies/langfuse/jobs/uHysbKH-developer-advocate-devrel
+- Backend Engineer, 70-130k EUR, 0.25-0.75% Equity, https://www.ycombinator.com/companies/langfuse/jobs/mnrdwla-backend-engineer
+- Product Engineer, 70-130k EUR, 0.25-0.75% Equity, https://www.ycombinator.com/companies/langfuse/jobs/aAvmoFB-product-engineer
+- Developer Advocate, 60-110k EUR, 0.25-0.5% Equity, https://www.ycombinator.com/companies/langfuse/jobs/uHysbKH-developer-advocate-devrel
 
 _More Info_
 

docker-compose.build.yml

@@ -5,6 +5,8 @@ services:
     build:
       dockerfile: ./web/Dockerfile
       context: .
+      args:
+        - NEXT_PUBLIC_LANGFUSE_CLOUD_REGION=${NEXT_PUBLIC_LANGFUSE_CLOUD_REGION}
     depends_on:
       - db
       - redis
@@ -20,11 +22,18 @@ services:
       - LANGFUSE_WORKER_HOST=${LANGFUSE_WORKER_HOST:-worker}
       - LANGFUSE_WORKER_PASSWORD=${LANGFUSE_WORKER_PASSWORD:-mybasicauthsecret}
     restart: always
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:3000/api/public/health"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
 
   worker:
     build:
       dockerfile: ./worker/Dockerfile
       context: .
+      args:
+        - NEXT_PUBLIC_LANGFUSE_CLOUD_REGION=${NEXT_PUBLIC_LANGFUSE_CLOUD_REGION}
     depends_on:
       - db
       - redis
@@ -41,6 +50,11 @@ services:
       - REDIS_AUTH=${REDIS_AUTH:-myredissecret}
       - LANGFUSE_WORKER_PASSWORD=${LANGFUSE_WORKER_PASSWORD:-mybasicauthsecret}
     restart: always
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:3030/api/health"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
 
   redis:
     image: redis:7.2.4

docker-compose.dev.yml

@@ -1,5 +1,3 @@
-version: "3.5"
-
 services:
   redis:
     image: redis:7.2.4
@@ -10,7 +8,7 @@ services:
       - 6379:6379
 
   db:
-    image: postgres
+    image: postgres:${POSTGRES_VERSION:-latest}
     restart: always
     command: ["postgres", "-c", "log_statement=all"]
     environment:

docker-compose.yml

@@ -1,8 +1,6 @@
-version: "3.5"
-
 services:
   langfuse-server:
-    image: ghcr.io/langfuse/langfuse:latest
+    image: langfuse/langfuse:2
     depends_on:
       db:
         condition: service_healthy

ee/package.json

@@ -33,7 +33,7 @@
     "@repo/eslint-config": "*",
     "@repo/typescript-config": "*",
     "@types/node": "^20.11.29",
-    "@typescript-eslint/parser": "^7.7.0",
+    "@typescript-eslint/parser": "^7.12.0",
     "eslint": "^8.57.0",
     "eslint-config-prettier": "^9.1.0",
     "eslint-config-standard": "^17.1.0",

ee/src/sso/index.ts

@@ -2,14 +2,16 @@ import { type Provider } from "next-auth/providers/index";
 import GoogleProvider from "next-auth/providers/google";
 import GitHubProvider from "next-auth/providers/github";
 import OktaProvider from "next-auth/providers/okta";
+import CognitoProvider from "next-auth/providers/cognito";
 import Auth0Provider from "next-auth/providers/auth0";
 import AzureADProvider from "next-auth/providers/azure-ad";
 import { isEeAvailable } from "..";
-import { prisma } from "@langfuse/shared/src/db";
+import { type SsoConfig, prisma } from "@langfuse/shared/src/db";
 import { encrypt, decrypt } from "@langfuse/shared/encryption";
 import { SsoProviderSchema } from "./types";
 import { type NextApiRequest, type NextApiResponse } from "next";
 import { env } from "../env";
+import { CustomSSOProvider } from "@langfuse/shared/src/server";
 
 // Local cache for SSO configurations
 let cachedSsoConfigs: {
@@ -25,14 +27,36 @@ let cachedSsoConfigs: {
 async function getSsoConfigs(): Promise<SsoProviderSchema[]> {
   if (!isEeAvailable) return [];
   const CACHE_TTL = 60 * 1000; // 1 minute
+  const DB_MAX_WAIT = 2000; // 2 seconds
+  const DB_TIMEOUT = 3000; // 3 seconds
 
   // Set/refresh the cache if it's empty or expired
   if (
     cachedSsoConfigs === null ||
     Date.now() - cachedSsoConfigs.timestamp > CACHE_TTL
   ) {
+    // findMany with custom timeout via $transaction
+    let dbConfigs: SsoConfig[] = [];
+    try {
+      dbConfigs = await prisma.$transaction(
+        async (prisma) => prisma.ssoConfig.findMany(),
+        {
+          maxWait: DB_MAX_WAIT,
+          timeout: DB_TIMEOUT,
+        }
+      );
+    } catch (e) {
+      // cache empty array to prevent repeated DB calls on error
+      cachedSsoConfigs = {
+        data: [],
+        timestamp: Date.now(),
+      };
+
+      // caught and logged in the caller
+      throw e;
+    }
+
     // transform into zod object
-    const dbConfigs = await prisma.ssoConfig.findMany();
     const parsedSsoConfigs = dbConfigs
       .map((v) => {
         try {
@@ -144,6 +168,18 @@ const dbToNextAuthProvider = (provider: SsoProviderSchema): Provider | null => {
       ...provider.authConfig,
       clientSecret: decrypt(provider.authConfig.clientSecret),
     });
+  else if (provider.authProvider === "cognito")
+    return CognitoProvider({
+      id: getAuthProviderIdForSsoConfig(provider), // use the domain as the provider id as we use domain-specific credentials
+      ...provider.authConfig,
+      clientSecret: decrypt(provider.authConfig.clientSecret),
+    });
+  else if (provider.authProvider === "custom")
+    return CustomSSOProvider({
+      id: getAuthProviderIdForSsoConfig(provider), // use the domain as the provider id as we use domain-specific credentials
+      ...provider.authConfig,
+      clientSecret: decrypt(provider.authConfig.clientSecret),
+    });
   else {
     // Type check to ensure we handle all providers
     // eslint-disable-next-line no-unused-vars

ee/src/sso/types.ts

@@ -64,11 +64,38 @@ export const AzureAdProviderSchema = base.extend({
     .nullish(),
 });
 
+export const CognitoProviderSchema = base.extend({
+  authProvider: z.literal("cognito"),
+  authConfig: z
+    .object({
+      clientId: z.string(),
+      clientSecret: z.string(),
+      issuer: z.string(),
+      allowDangerousEmailAccountLinking: z.boolean().optional().default(false),
+    })
+    .nullish(),
+});
+
+export const CustomProviderSchema = base.extend({
+  authProvider: z.literal("custom"),
+  authConfig: z
+    .object({
+      name: z.string(),
+      clientId: z.string(),
+      clientSecret: z.string(),
+      issuer: z.string(),
+      allowDangerousEmailAccountLinking: z.boolean().optional().default(false),
+    })
+    .nullish(),
+});
+
 export type GoogleProviderSchema = z.infer<typeof GoogleProviderSchema>;
 export type GithubProviderSchema = z.infer<typeof GithubProviderSchema>;
 export type Auth0ProviderSchema = z.infer<typeof Auth0ProviderSchema>;
 export type OktaProviderSchema = z.infer<typeof OktaProviderSchema>;
 export type AzureAdProviderSchema = z.infer<typeof AzureAdProviderSchema>;
+export type CognitoProviderSchema = z.infer<typeof CognitoProviderSchema>;
+export type CustomProviderSchema = z.infer<typeof CustomProviderSchema>;
 
 export const SsoProviderSchema = z.discriminatedUnion("authProvider", [
   GoogleProviderSchema,
@@ -76,6 +103,8 @@ export const SsoProviderSchema = z.discriminatedUnion("authProvider", [
   Auth0ProviderSchema,
   OktaProviderSchema,
   AzureAdProviderSchema,
+  CognitoProviderSchema,
+  CustomProviderSchema,
 ]);
 
 export type SsoProviderSchema = z.infer<typeof SsoProviderSchema>;

fern/apis/server/definition/commons.yml

@@ -111,7 +111,24 @@ types:
       totalCost:
         docs: USD total cost, defaults to input+output
         type: optional<double>
-
+  ScoreConfig:
+    docs: Configuration for a score
+    properties:
+      id: string
+      name: string
+      createdAt: datetime
+      updatedAt: datetime
+      projectId: string
+      dataType: ScoreDataType
+      isArchived: boolean
+      minValue: optional<double>
+      maxValue: optional<double>
+      categories: optional<list<ConfigCategory>>
+      description: optional<string>
+  ConfigCategory:
+    properties:
+      value: double
+      label: string
   Score:
     properties:
       id: string
@@ -209,9 +226,14 @@ types:
       - ARCHIVED
   ScoreSource:
     enum:
+      - ANNOTATION
       - API
-      - REVIEW
       - EVAL
+  ScoreDataType:
+    enum:
+      - NUMERIC
+      - BOOLEAN
+      - CATEGORICAL
 
 errors:
   Error:

fern/apis/server/definition/dataset-items.yml

@@ -31,7 +31,7 @@ types:
       sourceObservationId: optional<string>
       id:
         type: optional<string>
-        docs: Dataset items are upserted on their id
+        docs: Dataset items are upserted on their id. Id needs to be globally unique and cannot be reused across datasets.
       status:
         type: optional<commons.DatasetStatus>
         docs: Defaults to ACTIVE for newly created items

fern/apis/server/definition/ingestion.yml

@@ -7,28 +7,57 @@ service:
   base-path: /api/public
   endpoints:
     batch:
-      docs: Batched ingestion for Langfuse Tracing
+      docs: |
+        Batched ingestion for Langfuse Tracing. If you want to use tracing via the API, such as to build your own Langfuse client implementation, this is the only API route you need to implement.
+
+        Notes:
+        - Batch sizes are limited to 3.5 MB in total. You need to adjust the number of events per batch accordingly.
+        - The API does not return a 4xx status code for input errors. Instead, it responds with a 207 status code, which includes a list of the encountered errors.
       method: POST
       path: /ingestion
       request:
         name: IngestionRequest
         body:
           properties:
-            batch: list<IngestionEvent>
-      response: IngestionResponse # will be reportet as 200 response, but endpoint returns 207
+            batch:
+              type: list<IngestionEvent>
+              docs: "Batch of tracing events to be ingested. Discriminated by attribute `type`."
+            metadata:
+              type: optional<unknown>
+              docs: Optional. Metadata field used by the Langfuse SDKs for debugging.
+      response:
+        type: IngestionResponse
+        status-code: 207
 
 types:
   IngestionEvent:
     discriminant: "type"
     union:
-      trace-create: TraceEvent
-      score-create: ScoreEvent
-      event-create: CreateEventEvent
-      generation-create: CreateGenerationEvent
-      generation-update: UpdateGenerationEvent
-      span-create: CreateSpanEvent
-      span-update: UpdateSpanEvent
-      sdk-log: SDKLogEvent
+      trace-create:
+        type: TraceEvent
+        docs: Creates a new trace. Upserts on id for updates if trace with id exists.
+      score-create:
+        type: ScoreEvent
+        docs: Creates a new score. Upserts on id for updates if score with id exists.
+      span-create:
+        type: CreateSpanEvent
+        docs: Creates a new span.
+      span-update:
+        type: UpdateSpanEvent
+        docs: Updates span based on id.
+      generation-create:
+        type: CreateGenerationEvent
+        docs: Creates a new generation.
+      generation-update:
+        type: UpdateGenerationEvent
+        docs: Updates a generation based on id.
+      event-create:
+        type: CreateEventEvent
+        docs: Creates an event.
+
+      sdk-log:
+        type: SDKLogEvent
+        docs: Langfuse SDKs only, used for debugging purposes.
 
       # both are legacy
       observation-create:
@@ -162,9 +191,15 @@ types:
 
   BaseEvent:
     properties:
-      id: string
-      timestamp: string
-      metadata: unknown
+      id:
+        type: string
+        docs: UUID v4 that identifies the event
+      timestamp:
+        type: string
+        docs: "Datetime (ISO 8601) of event creation in client. Should be as close to actual event creation in client as possible, this timestamp will be used for ordering of events in future release. Resolution: milliseconds (required), microseconds (optimal)."
+      metadata:
+        type: optional<unknown>
+        docs: Optional. Metadata field used by the Langfuse SDKs for debugging.
 
   TraceEvent:
     extends: BaseEvent

fern/apis/server/definition/metrics.yml

@@ -48,12 +48,25 @@ types:
       date: date
       countTraces: integer
       countObservations: integer
-      totalCost: double
+      totalCost:
+        type: double
+        docs: Total model cost in USD
       usage: list<UsageByModel>
   UsageByModel:
     docs: Daily usage of a given model. Usage corresponds to the unit set for the specific model (e.g. tokens).
     properties:
       model: optional<string>
-      inputUsage: integer
-      outputUsage: integer
-      totalUsage: integer
+      inputUsage:
+        type: integer
+        docs: Total number of generation input units (e.g. tokens)
+      outputUsage:
+        type: integer
+        docs: Total number of generation output units (e.g. tokens)
+      totalUsage:
+        type: integer
+        docs: Total number of generation total units (e.g. tokens)
+      countTraces: integer
+      countObservations: integer
+      totalCost:
+        type: double
+        docs: Total model cost in USD

fern/apis/server/definition/prompts.yml

@@ -70,14 +70,24 @@ types:
       name: string
       prompt: list<ChatMessage>
       config: optional<unknown>
-      labels: optional<list<string>>
+      labels:
+        type: optional<list<string>>
+        docs: List of deployment labels of this prompt version.
+      tags:
+        type: optional<list<string>>
+        docs: List of tags to apply to all versions of this prompt.
 
   CreateTextPromptRequest:
     properties:
       name: string
       prompt: string
       config: optional<unknown>
-      labels: optional<list<string>>
+      labels:
+        type: optional<list<string>>
+        docs: List of deployment labels of this prompt version.
+      tags:
+        type: optional<list<string>>
+        docs: List of tags to apply to all versions of this prompt.
 
   Prompt:
     union:
@@ -89,7 +99,12 @@ types:
       name: string
       version: integer
       config: unknown
-      labels: list<string>
+      labels:
+        type: list<string>
+        docs: List of deployment labels of this prompt version.
+      tags:
+        type: list<string>
+        docs: List of tags. Used to filter via UI and API. The same across versions of a prompt.
 
   ChatMessage:
     properties:

fern/apis/server/definition/score-configs.yml

@@ -0,0 +1,36 @@
+# yaml-language-server: $schema=https://raw.githubusercontent.com/fern-api/fern/main/fern.schema.json
+imports:
+  pagination: ./utils/pagination.yml
+  commons: ./commons.yml
+service:
+  auth: true
+  base-path: /api/public
+  endpoints:
+    get:
+      docs: Get all score configs
+      method: GET
+      path: /score-configs
+      request:
+        name: GetScoreConfigsRequest
+        query-parameters:
+          page:
+            type: optional<integer>
+            docs: Page number, starts at 1.
+          limit:
+            type: optional<integer>
+            docs: Limit of items per page. If you encounter api issues due to too large page sizes, try to reduce the limit.
+      response: ScoreConfigs
+    get-by-id:
+      docs: Get a score config
+      method: GET
+      path: /score-configs/{configId}
+      path-parameters:
+        configId:
+          type: string
+          docs: The unique langfuse identifier of a score config
+      response: commons.ScoreConfig
+types:
+  ScoreConfigs:
+    properties:
+      data: list<commons.ScoreConfig>
+      meta: pagination.MetaResponse

fern/apis/server/definition/score.yml

@@ -31,7 +31,7 @@ service:
             type: optional<datetime>
             docs: Retrieve only scores newer than this datetime (ISO 8601).
           source:
-            type: commons.ScoreSource
+            type: optional<commons.ScoreSource>
             docs: Retrieve only scores from a specific source.
           operator:
             type: optional<string>
@@ -39,6 +39,9 @@ service:
           value:
             type: optional<double>
             docs: Retrieve only scores with <operator> value.
+          scoreIds:
+            type: optional<string>
+            docs: Comma-separated list of score IDs to limit the results to.
       response: Scores
     get-by-id:
       docs: Get a score

fern/apis/server/definition/sessions.yml

@@ -7,7 +7,7 @@ service:
   base-path: /api/public
   endpoints:
     get:
-      docs: Get a session
+      docs: Get a session. Please note that `traces` on this endpoint are not paginated, if you plan to fetch large sessions, consider `GET /api/public/traces?sessionId=<sessionId>`
       method: GET
       path: /sessions/{sessionId}
       path-parameters:

fern/apis/server/definition/trace.yml

@@ -30,6 +30,7 @@ service:
             docs: Limit of items per page. If you encounter api issues due to too large page sizes, try to reduce the limit.
           userId: optional<string>
           name: optional<string>
+          sessionId: optional<string>
           fromTimestamp:
             type: optional<datetime>
             docs: Retrieve only traces newer than this datetime (ISO 8601).

fern/fern.config.json

@@ -1,4 +1,4 @@
 {
   "organization": "finto",
-  "version": "0.21.0"
+  "version": "0.30.3"
 }

package.json

@@ -1,6 +1,6 @@
 {
   "name": "langfuse",
-  "version": "2.45.0",
+  "version": "2.57.0",
   "author": "engineering@langfuse.com",
   "license": "MIT",
   "private": true,
@@ -17,26 +17,36 @@
     "db:seed": "turbo run db:seed",
     "db:seed:examples": "turbo run db:seed:examples",
     "nuke": "bash ./scripts/nuke.sh",
-    "dx": "pnpm i && pnpm --filter=shared run db:reset && pnpm --filter=shared run db:seed:examples && pnpm run dev",
+    "dx": "pnpm i && pnpm run infra:dev:up && pnpm --filter=shared run db:reset && pnpm --filter=shared run db:seed:examples && pnpm run dev",
+    "dx-f": "pnpm i && pnpm run infra:dev:up && pnpm --filter=shared run db:reset -f && pnpm --filter=shared run db:seed:examples && pnpm run dev",
     "build": "turbo run build",
     "start": "turbo run start",
     "dev": "turbo run dev",
     "lint": "turbo run lint",
     "test": "turbo run test",
     "models:migrate": "turbo run models:migrate",
-    "release": "dotenv -e ../.env -- release-it"
+    "release": "dotenv -e ../.env -- release-it",
+    "prepare": "husky"
   },
   "devDependencies": {
     "@release-it/bumper": "^6.0.1",
+    "braces": "3.0.3",
     "dotenv-cli": "^7.4.2",
+    "husky": "^9.0.11",
     "prettier": "^3.2.5",
-    "release-it": "^17.2.1",
+    "release-it": "^17.3.0",
     "turbo": "^1.13.3"
   },
   "release-it": {
     "git": {
       "commitMessage": "chore: release v${version}",
-      "tagName": "v${version}"
+      "tagName": "v${version}",
+      "commitArgs": [
+        "--no-verify"
+      ],
+      "pushArgs": [
+        "--no-verify"
+      ]
     },
     "plugins": {
       "@release-it/bumper": {

packages/config-eslint/package.json

@@ -9,7 +9,7 @@
   ],
   "devDependencies": {
     "@typescript-eslint/eslint-plugin": "^7.1.0",
-    "@typescript-eslint/parser": "^7.7.0",
+    "@typescript-eslint/parser": "^7.12.0",
     "@vercel/style-guide": "^6.0.0",
     "eslint-config-next": "^14.2.3",
     "eslint-config-prettier": "^9.1.0",

packages/shared/package.json

@@ -17,9 +17,9 @@
       "import": "./dist/src/db.js",
       "require": "./dist/src/db.js"
     },
-    "./src/server/auth": {
-      "import": "./dist/src/server/auth.js",
-      "require": "./dist/src/server/auth.js"
+    "./src/server": {
+      "import": "./dist/src/server/index.js",
+      "require": "./dist/src/server/index.js"
     },
     "./encryption": {
       "import": "./dist/src/encryption/index.js",
@@ -44,14 +44,22 @@
     "seed": "ts-node -r tsconfig-paths/register -r dotenv/config --compiler-options {\"module\":\"CommonJS\"} prisma/seed.ts"
   },
   "dependencies": {
+    "@aws-sdk/client-s3": "^3.550.0",
+    "@aws-sdk/lib-storage": "^3.568.0",
+    "@aws-sdk/s3-request-presigner": "^3.554.0",
     "@langchain/anthropic": "^0.1.21",
     "@langchain/core": "^0.1.61",
     "@langchain/openai": "^0.0.33",
     "@prisma/client": "^5.13.0",
+    "@react-email/components": "^0.0.19",
+    "@react-email/render": "^0.0.15",
     "@types/bcryptjs": "^2.4.6",
     "bcryptjs": "^2.4.3",
     "kysely": "^0.27.3",
-    "langchain": "^0.1.36",
+    "langchain": "^0.2.6",
+    "lodash": "^4.17.21",
+    "next-auth": "^4.24.7",
+    "nodemailer": "^6.9.13",
     "prisma-extension-kysely": "^2.1.0",
     "zod": "^3.22.4",
     "zod-to-json-schema": "^3.22.5"
@@ -59,17 +67,18 @@
   "devDependencies": {
     "@repo/eslint-config": "*",
     "@repo/typescript-config": "*",
-    "@types/lodash": "^4.17.4",
+    "@types/lodash": "^4.17.5",
     "@types/node": "^20.11.29",
+    "@types/nodemailer": "^6.4.15",
     "@types/pg": "^8.11.6",
     "@types/uuid": "^9.0.8",
-    "@typescript-eslint/parser": "^7.7.0",
+    "@typescript-eslint/parser": "^7.12.0",
     "eslint": "^8.57.0",
     "eslint-config-prettier": "^9.1.0",
     "eslint-config-standard": "^17.1.0",
     "eslint-plugin-prettier": "^5.1.3",
     "kysely-codegen": "^0.11.0",
-    "nodemon": "^3.0.3",
+    "nodemon": "^3.1.3",
     "prettier": "^3.2.5",
     "prisma": "^5.13.0",
     "prisma-erd-generator": "^1.11.2",
@@ -78,5 +87,9 @@
     "tsc-watch": "^6.2.0",
     "typescript": "^5.4.5",
     "vitest": "^1.5.3"
+  },
+  "peerDependencies": {
+    "@types/react": "^18.2.79",
+    "react": "^18.0.0"
   }
 }

packages/shared/prisma/generated/types.ts

@@ -25,22 +25,17 @@ export const ObservationLevel = {
 } as const;
 export type ObservationLevel = (typeof ObservationLevel)[keyof typeof ObservationLevel];
 export const ScoreSource = {
+    ANNOTATION: "ANNOTATION",
     API: "API",
-    REVIEW: "REVIEW",
     EVAL: "EVAL"
 } as const;
 export type ScoreSource = (typeof ScoreSource)[keyof typeof ScoreSource];
-export const PricingUnit = {
-    PER_1000_TOKENS: "PER_1000_TOKENS",
-    PER_1000_CHARS: "PER_1000_CHARS"
+export const ScoreDataType = {
+    CATEGORICAL: "CATEGORICAL",
+    NUMERIC: "NUMERIC",
+    BOOLEAN: "BOOLEAN"
 } as const;
-export type PricingUnit = (typeof PricingUnit)[keyof typeof PricingUnit];
-export const TokenType = {
-    PROMPT: "PROMPT",
-    COMPLETION: "COMPLETION",
-    TOTAL: "TOTAL"
-} as const;
-export type TokenType = (typeof TokenType)[keyof typeof TokenType];
+export type ScoreDataType = (typeof ScoreDataType)[keyof typeof ScoreDataType];
 export const DatasetStatus = {
     ACTIVE: "ACTIVE",
     ARCHIVED: "ARCHIVED"
@@ -103,6 +98,21 @@ export type AuditLog = {
     before: string | null;
     after: string | null;
 };
+export type BatchExport = {
+    id: string;
+    created_at: Generated<Timestamp>;
+    updated_at: Generated<Timestamp>;
+    project_id: string;
+    user_id: string;
+    finished_at: Timestamp | null;
+    expires_at: Timestamp | null;
+    name: string;
+    status: string;
+    query: unknown;
+    format: string;
+    url: string | null;
+    log: string | null;
+};
 export type CronJobs = {
     name: string;
     last_run: Timestamp | null;
@@ -157,6 +167,7 @@ export type EvalTemplate = {
     version: number;
     prompt: string;
     model: string;
+    provider: string;
     model_params: unknown;
     vars: Generated<string[]>;
     output_schema: unknown;
@@ -204,8 +215,12 @@ export type LlmApiKeys = {
     created_at: Generated<Timestamp>;
     updated_at: Generated<Timestamp>;
     provider: string;
+    adapter: string;
     display_secret_key: string;
     secret_key: string;
+    base_url: string | null;
+    custom_models: Generated<string[]>;
+    with_default_models: Generated<boolean>;
     project_id: string;
 };
 export type MembershipInvitation = {
@@ -246,8 +261,10 @@ export type Observation = {
     status_message: string | null;
     version: string | null;
     created_at: Generated<Timestamp>;
+    updated_at: Generated<Timestamp>;
     model: string | null;
     internal_model: string | null;
+    internal_model_id: string | null;
     modelParameters: unknown | null;
     input: unknown | null;
     output: unknown | null;
@@ -258,6 +275,9 @@ export type Observation = {
     input_cost: string | null;
     output_cost: string | null;
     total_cost: string | null;
+    calculated_input_cost: string | null;
+    calculated_output_cost: string | null;
+    calculated_total_cost: string | null;
     completion_start_time: Timestamp | null;
     prompt_id: string | null;
 };
@@ -303,14 +323,6 @@ export type PosthogIntegration = {
     enabled: boolean;
     created_at: Generated<Timestamp>;
 };
-export type Pricing = {
-    id: string;
-    model_name: string;
-    pricing_unit: Generated<PricingUnit>;
-    price: string;
-    currency: Generated<string>;
-    token_type: TokenType;
-};
 export type Project = {
     id: string;
     created_at: Generated<Timestamp>;
@@ -351,6 +363,24 @@ export type Score = {
     comment: string | null;
     trace_id: string;
     observation_id: string | null;
+    config_id: string | null;
+    string_value: string | null;
+    created_at: Generated<Timestamp>;
+    updated_at: Generated<Timestamp>;
+    data_type: Generated<ScoreDataType>;
+};
+export type ScoreConfig = {
+    id: string;
+    created_at: Generated<Timestamp>;
+    updated_at: Generated<Timestamp>;
+    project_id: string;
+    name: string;
+    data_type: ScoreDataType;
+    is_archived: Generated<boolean>;
+    min_value: number | null;
+    max_value: number | null;
+    categories: unknown | null;
+    description: string | null;
 };
 export type Session = {
     id: string;
@@ -431,6 +461,7 @@ export type DB = {
     Account: Account;
     api_keys: ApiKey;
     audit_logs: AuditLog;
+    batch_exports: BatchExport;
     cron_jobs: CronJobs;
     dataset_items: DatasetItem;
     dataset_run_items: DatasetRunItems;
@@ -446,10 +477,10 @@ export type DB = {
     observations: Observation;
     observations_view: ObservationView;
     posthog_integrations: PosthogIntegration;
-    pricings: Pricing;
     project_memberships: ProjectMembership;
     projects: Project;
     prompts: Prompt;
+    score_configs: ScoreConfig;
     scores: Score;
     Session: Session;
     sso_configs: SsoConfig;

packages/shared/prisma/migrations/20240523142425_score_config_add_table/migration.sql

@@ -0,0 +1,42 @@
+-- CreateEnum
+CREATE TYPE "ScoreDataType" AS ENUM ('CATEGORICAL', 'NUMERIC');
+
+-- AlterTable
+ALTER TABLE "scores" ADD COLUMN     "config_id" TEXT,
+ADD COLUMN     "data_type" "ScoreDataType" NOT NULL DEFAULT 'NUMERIC',
+ADD COLUMN     "string_value" TEXT;
+
+-- CreateTable
+CREATE TABLE "score_configs" (
+    "id" TEXT NOT NULL,
+    "created_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    "updated_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    "project_id" TEXT NOT NULL,
+    "name" TEXT NOT NULL,
+    "data_type" "ScoreDataType" NOT NULL,
+    "is_archived" BOOLEAN NOT NULL DEFAULT false,
+    "min_value" DOUBLE PRECISION,
+    "max_value" DOUBLE PRECISION,
+    "categories" JSONB,
+    "description" TEXT,
+
+    CONSTRAINT "score_configs_pkey" PRIMARY KEY ("id")
+);
+
+-- CreateIndex
+CREATE INDEX "score_configs_data_type_idx" ON "score_configs"("data_type");
+
+-- CreateIndex
+CREATE INDEX "score_configs_is_archived_idx" ON "score_configs"("is_archived");
+
+-- CreateIndex
+CREATE INDEX "score_configs_project_id_idx" ON "score_configs"("project_id");
+
+-- CreateIndex
+CREATE INDEX "score_configs_categories_idx" ON "score_configs"("categories");
+
+-- CreateIndex
+CREATE UNIQUE INDEX "score_configs_id_project_id_key" ON "score_configs"("id", "project_id");
+
+-- AddForeignKey
+ALTER TABLE "score_configs" ADD CONSTRAINT "score_configs_project_id_fkey" FOREIGN KEY ("project_id") REFERENCES "projects"("id") ON DELETE CASCADE ON UPDATE CASCADE;

packages/shared/prisma/migrations/20240523142524_scores_add_config_id_idx/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "scores_config_id_idx" ON "scores"("config_id");

packages/shared/prisma/migrations/20240523142610_scores_add_fk_scores_config_id/migration.sql

@@ -0,0 +1,2 @@
+-- AddForeignKey
+ALTER TABLE "scores" ADD CONSTRAINT "scores_config_id_fkey" FOREIGN KEY ("config_id") REFERENCES "score_configs"("id") ON DELETE SET NULL ON UPDATE CASCADE;

packages/shared/prisma/migrations/20240524154058_scores_source_enum_add_annotation/migration.sql

@@ -0,0 +1,2 @@
+-- AlterEnum
+ALTER TYPE "ScoreSource" ADD VALUE 'ANNOTATION';

packages/shared/prisma/migrations/20240524156058_scores_source_backfill_annotation_for_review/migration.sql

@@ -0,0 +1,4 @@
+-- Backfill the scores source for 'REVIEW' to be 'ANNOTATION'
+UPDATE "scores"
+SET "source" = 'ANNOTATION'::"ScoreSource"
+WHERE "source" = 'REVIEW'::"ScoreSource";

packages/shared/prisma/migrations/20240524165931_scores_source_enum_drop_review/migration.sql

@@ -0,0 +1,14 @@
+/*
+  Warnings:
+
+  - The values [REVIEW] on the enum `ScoreSource` will be removed. If these variants are still used in the database, this will fail.
+
+*/
+-- AlterEnum
+BEGIN;
+CREATE TYPE "ScoreSource_new" AS ENUM ('ANNOTATION', 'API', 'EVAL');
+ALTER TABLE "scores" ALTER COLUMN "source" TYPE "ScoreSource_new" USING ("source"::text::"ScoreSource_new");
+ALTER TYPE "ScoreSource" RENAME TO "ScoreSource_old";
+ALTER TYPE "ScoreSource_new" RENAME TO "ScoreSource";
+DROP TYPE "ScoreSource_old";
+COMMIT;

packages/shared/prisma/migrations/20240524190433_job_executions_add_fk_index_config_id/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "job_executions_job_configuration_id_idx" ON "job_executions"("job_configuration_id");

packages/shared/prisma/migrations/20240524190434_job_executions_add_fk_index_score_id/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "job_executions_job_output_score_id_idx" ON "job_executions"("job_output_score_id");

packages/shared/prisma/migrations/20240524190435_job_executions_add_fk_index_trace_id/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "job_executions_job_input_trace_id_idx" ON "job_executions"("job_input_trace_id");

packages/shared/prisma/migrations/20240524190436_job_executions_index_created_at/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "job_executions_created_at_idx" ON "job_executions"("created_at");

packages/shared/prisma/migrations/20240528214726_add_cursor_new_columns_observations/migration.sql

@@ -0,0 +1,2 @@
+-- AlterTable
+ALTER TABLE "observations" ADD COLUMN     "updated_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP;

packages/shared/prisma/migrations/20240528214727_add_cursor_new_columns_scores/migration.sql

@@ -0,0 +1,4 @@
+
+-- AlterTable
+ALTER TABLE "scores" ADD COLUMN     "created_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+ADD COLUMN     "updated_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP;

packages/shared/prisma/migrations/20240528214728_add_cursor_index_01/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "audit_logs_updated_at_idx" ON "audit_logs"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_02/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "dataset_items_created_at_idx" ON "dataset_items"("created_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_03/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "dataset_items_updated_at_idx" ON "dataset_items"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_04/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "dataset_run_items_created_at_idx" ON "dataset_run_items"("created_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_05/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "dataset_run_items_updated_at_idx" ON "dataset_run_items"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_06/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "dataset_runs_created_at_idx" ON "dataset_runs"("created_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_07/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "dataset_runs_updated_at_idx" ON "dataset_runs"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_08/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "datasets_created_at_idx" ON "datasets"("created_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_09/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "datasets_updated_at_idx" ON "datasets"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_10/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "job_executions_updated_at_idx" ON "job_executions"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_11/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "observations_updated_at_idx" ON "observations"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_12/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "prompts_created_at_idx" ON "prompts"("created_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_13/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "prompts_updated_at_idx" ON "prompts"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_14/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "score_configs_created_at_idx" ON "score_configs"("created_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_15/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "score_configs_updated_at_idx" ON "score_configs"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_16/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "scores_created_at_idx" ON "scores"("created_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_17/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "scores_updated_at_idx" ON "scores"("updated_at");

packages/shared/prisma/migrations/20240528214728_add_cursor_index_18/migration.sql

@@ -0,0 +1,3 @@
+
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "trace_sessions_updated_at_idx" ON "trace_sessions"("updated_at");

packages/shared/prisma/migrations/20240603212024_dataset_items_add_index_source_trace_id/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "dataset_items_source_trace_id_idx" ON "dataset_items" USING HASH ("source_trace_id");

packages/shared/prisma/migrations/20240604133338_scores_add_index_name/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX "scores_project_id_name_idx" ON "scores"("project_id", "name");

packages/shared/prisma/migrations/20240604133339_score_data_type_add_boolean/migration.sql

@@ -0,0 +1,2 @@
+-- AlterEnum
+ALTER TYPE "ScoreDataType" ADD VALUE 'BOOLEAN';

packages/shared/prisma/migrations/20240606093356_drop_unused_pricings_table/migration.sql

@@ -0,0 +1,14 @@
+/*
+  Warnings:
+
+  - You are about to drop the `pricings` table. If the table is not empty, all the data it contains will be lost.
+
+*/
+-- DropTable
+DROP TABLE "pricings";
+
+-- DropEnum
+DROP TYPE "PricingUnit";
+
+-- DropEnum
+DROP TYPE "TokenType";

packages/shared/prisma/migrations/20240606133011_remove_trace_fkey_datasetrunitems/migration.sql

@@ -0,0 +1,3 @@
+-- DropForeignKey
+ALTER TABLE "dataset_run_items" DROP CONSTRAINT "dataset_run_items_trace_id_fkey";
+ALTER TABLE "dataset_run_items" DROP CONSTRAINT "dataset_run_items_observation_id_fkey";

packages/shared/prisma/migrations/20240607090858_pricings_add_latest_gemini_models/migration.sql

@@ -0,0 +1,8 @@
+-- does not include pricing yet, will be added as soon as it is calculated at ingestion time
+
+-- remove model if added by faulty migration, context: https://github.com/langfuse/langfuse/issues/2266
+DELETE FROM models WHERE id in ('clx30djsn0000w9mzebiv41we', 'clx30hkrx0000w9mz7lqi0ial');
+
+INSERT INTO "models" ("id", "model_name", "match_pattern", "unit") VALUES ('clx30djsn0000w9mzebiv41we', 'gemini-1.5-flash', '(?i)^(gemini-1.5-flash)(@[a-zA-Z0-9]+)?$', 'CHARACTERS');
+
+INSERT INTO "models" ("id", "model_name", "match_pattern", "unit") VALUES ('clx30hkrx0000w9mz7lqi0ial', 'gemini-1.5-pro', '(?i)^(gemini-1.5-pro)(@[a-zA-Z0-9]+)?$', 'CHARACTERS');

packages/shared/prisma/migrations/20240607212419_model_price_anthropic_via_google_vertex/migration.sql

@@ -0,0 +1,7 @@
+-- Google Vertex uses @ to separate model name and version
+
+UPDATE "models" SET "match_pattern" = '(?i)^(claude-3-haiku(-|@)?20240307)$' WHERE "id" = 'cltr0w45b000008k1407o9qv1';
+
+UPDATE "models" SET "match_pattern" = '(?i)^(claude-3-opus(-|@)?20240229)$' WHERE "id" = 'cltgy0iuw000008le3vod1hhy';
+
+UPDATE "models" SET "match_pattern" = '(?i)^(claude-3-sonnet(-|@)?20240229)$' WHERE "id" = 'cltgy0pp6000108le56se7bl3';

packages/shared/prisma/migrations/20240611105521_llm_api_keys_custom_endpoints/migration.sql

@@ -0,0 +1,24 @@
+BEGIN;
+
+  ALTER TABLE "llm_api_keys" 
+    ADD COLUMN "base_url" TEXT,
+    ADD COLUMN "adapter" TEXT,
+    ADD COLUMN "custom_models" TEXT[] NOT NULL DEFAULT '{}'::TEXT[],
+    ADD COLUMN "with_default_models" BOOLEAN NOT NULL DEFAULT true;
+
+  UPDATE "llm_api_keys"
+    SET "adapter" = "provider";
+
+  ALTER TABLE "llm_api_keys" 
+    ALTER COLUMN "adapter" SET NOT NULL;
+
+  ALTER TABLE "eval_templates"
+    ADD COLUMN "provider" TEXT;
+  
+  UPDATE "eval_templates"
+    SET "provider" = 'openai';
+  
+  ALTER TABLE "eval_templates"
+    ALTER COLUMN "provider" SET NOT NULL;
+
+COMMIT;

packages/shared/prisma/migrations/20240611113517_backfill_manual_score_configs/migration.sql

@@ -0,0 +1,51 @@
+-- Create manual-score config for any project with manual scores and link config_id to scores
+BEGIN;
+WITH project_configs AS (
+INSERT INTO score_configs (id,
+		project_id,
+		name,
+		data_type,
+		is_archived,
+		min_value,
+		max_value,
+		description)
+SELECT
+	md5(random()::text || clock_timestamp()::text || s.project_id::text)::uuid AS id,
+	s.project_id,
+	'manual-score',
+	'NUMERIC',
+	FALSE,
+	- 1,
+	1,
+	'Langfuse legacy annotation score.'
+FROM ( SELECT DISTINCT
+		project_id
+	FROM
+		scores
+	WHERE
+		name = 'manual-score'
+		AND config_id IS NULL
+		AND source = 'ANNOTATION') s
+WHERE
+	NOT EXISTS (
+		SELECT
+			1
+		FROM
+			score_configs sc
+		WHERE
+			sc.name = 'manual-score'
+			AND sc.project_id = s.project_id)
+	RETURNING
+		id,
+		project_id
+)
+UPDATE
+	scores
+SET
+	config_id = pc.id
+FROM
+	project_configs pc
+WHERE
+	scores.project_id = pc.project_id
+	AND scores.name = 'manual-score';
+COMMIT;

packages/shared/prisma/migrations/20240612101858_add_index_observations_project_id_prompt_id/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "observations_project_id_prompt_id_idx" ON "observations"("project_id", "prompt_id");

packages/shared/prisma/migrations/20240617094803_observations_remove_prompt_fk_constraint/migration.sql

@@ -0,0 +1,2 @@
+-- DropForeignKey
+ALTER TABLE "observations" DROP CONSTRAINT "observations_prompt_id_fkey";

packages/shared/prisma/migrations/20240618134129_add_batch_exports_table/migration.sql

@@ -0,0 +1,27 @@
+-- CreateTable
+CREATE TABLE "batch_exports" (
+    "id" TEXT NOT NULL,
+    "created_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    "updated_at" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    "project_id" TEXT NOT NULL,
+    "user_id" TEXT NOT NULL,
+    "finished_at" TIMESTAMP(3),
+    "expires_at" TIMESTAMP(3),
+    "name" TEXT NOT NULL,
+    "status" TEXT NOT NULL,
+    "query" JSONB NOT NULL,
+    "format" TEXT NOT NULL,
+    "url" TEXT,
+    "log" TEXT,
+
+    CONSTRAINT "batch_exports_pkey" PRIMARY KEY ("id")
+);
+
+-- CreateIndex
+CREATE INDEX "batch_exports_project_id_user_id_idx" ON "batch_exports"("project_id", "user_id");
+
+-- CreateIndex
+CREATE INDEX "batch_exports_status_idx" ON "batch_exports"("status");
+
+-- AddForeignKey
+ALTER TABLE "batch_exports" ADD CONSTRAINT "batch_exports_project_id_fkey" FOREIGN KEY ("project_id") REFERENCES "projects"("id") ON DELETE CASCADE ON UPDATE CASCADE;

packages/shared/prisma/migrations/20240618164950_drop_observations_parent_observation_id_idx/migration.sql

@@ -0,0 +1,2 @@
+-- DropIndex
+DROP INDEX CONCURRENTLY IF EXISTS "observations_parent_observation_id_idx";

packages/shared/prisma/migrations/20240618164951_drop_observations_updated_at_idx/migration.sql

@@ -0,0 +1,2 @@
+-- DropIndex
+DROP INDEX CONCURRENTLY IF EXISTS "observations_updated_at_idx";

packages/shared/prisma/migrations/20240618164952_drop_scores_updated_at_idx/migration.sql

@@ -0,0 +1,2 @@
+-- DropIndex
+DROP INDEX CONCURRENTLY IF EXISTS "scores_updated_at_idx";

packages/shared/prisma/migrations/20240618164953_drop_traces_external_id_idx/migration.sql

@@ -0,0 +1,2 @@
+-- DropIndex
+DROP INDEX CONCURRENTLY IF EXISTS "traces_external_id_idx";

packages/shared/prisma/migrations/20240618164954_drop_traces_release_idx/migration.sql

@@ -0,0 +1,2 @@
+-- DropIndex
+DROP INDEX CONCURRENTLY IF EXISTS "traces_release_idx";

packages/shared/prisma/migrations/20240618164955_drop_traces_updated_at_idx/migration.sql

@@ -0,0 +1,2 @@
+-- DropIndex
+DROP INDEX CONCURRENTLY IF EXISTS "traces_updated_at_idx";

packages/shared/prisma/migrations/20240618164956_create_traces_project_id_timestamp_idx/migration.sql

@@ -0,0 +1,2 @@
+-- CreateIndex
+CREATE INDEX CONCURRENTLY "traces_project_id_timestamp_idx" ON "traces"("project_id", "timestamp");

packages/shared/prisma/migrations/20240624133412_models_add_anthropic_3_5_sonnet/migration.sql

@@ -0,0 +1,15 @@
+INSERT INTO models (
+  id,
+  project_id,
+  model_name,
+  match_pattern,
+  start_date,
+  input_price,
+  output_price,
+  total_price,
+  unit,
+  tokenizer_id
+)
+VALUES
+  -- add 3.5 sonnet model
+  ('clxt0n0m60000pumz1j5b7zsf', NULL, 'claude-3-5-sonnet-20240620', '(?i)^(claude-3-5-sonnet(-|@)?20240620)$', NULL, 0.000003, 0.000015, NULL, 'TOKENS', 'claude')

packages/shared/prisma/migrations/20240625103957_observations_add_calculated_cost_columns/migration.sql

@@ -0,0 +1,5 @@
+-- AlterTable
+ALTER TABLE "observations" ADD COLUMN     "calculated_input_cost" DECIMAL(65,30),
+ADD COLUMN     "calculated_output_cost" DECIMAL(65,30),
+ADD COLUMN     "calculated_total_cost" DECIMAL(65,30),
+ADD COLUMN     "internal_model_id" TEXT;

packages/shared/prisma/migrations/20240625103958_fix_model_match_gpt4_vision/migration.sql

@@ -0,0 +1,17 @@
+/*
+Previously, the pattern did not allow for model specifications like "1106" or any other 4-digit block. 
+This led to missing models on the generation-update call where the exact model name that was used was provided (including the 4-digit block).
+The new pattern allows for:
+
+- "gpt-4-vision-preview" as set on generation-create
+- "gpt-4-1106-vision-preview" as set on generation-update
+
+*/
+
+UPDATE
+	"models"
+SET
+	"match_pattern" = '(?i)^(gpt-4(-\d{4})?-vision-preview)$'
+WHERE
+	"id" = 'clrkvx5gp000108juaogs54ea'
+	AND "model_name" = 'gpt-4-turbo-vision';

packages/shared/prisma/schema.prisma

@@ -114,6 +114,8 @@ model Project {
     LlmApiKeys         LlmApiKeys[]
     PosthogIntegration PosthogIntegration[]
     Score              Score[]
+    scoreConfig        ScoreConfig[]
+    BatchExport        BatchExport[]
 
     @@map("projects")
 }
@@ -143,9 +145,13 @@ model LlmApiKeys {
     createdAt DateTime @default(now()) @map("created_at")
     updatedAt DateTime @default(now()) @updatedAt @map("updated_at")
 
-    provider         String
-    displaySecretKey String @map("display_secret_key")
-    secretKey        String @map("secret_key")
+    provider          String
+    adapter           String // This controls the interface that is used to connect with the LLM, e.g. 'openai' or 'anthropic'
+    displaySecretKey  String   @map("display_secret_key")
+    secretKey         String   @map("secret_key")
+    baseURL           String?  @map("base_url")
+    customModels      String[] @default([]) @map("custom_models")
+    withDefaultModels Boolean  @default(true) @map("with_default_models")
 
     projectId String  @map("project_id")
     project   Project @relation(fields: [projectId], references: [id], onDelete: Cascade)
@@ -205,6 +211,7 @@ model TraceSession {
     @@id([id, projectId])
     @@index([projectId])
     @@index([createdAt])
+    @@index([updatedAt])
     @@map("trace_sessions")
 }
 
@@ -231,19 +238,16 @@ model Trace {
     createdAt  DateTime      @default(now()) @map("created_at")
     updatedAt  DateTime      @default(now()) @updatedAt @map("updated_at")
 
-    DatasetRunItems DatasetRunItems[]
-    DatasetItem     DatasetItem[]
-    JobExecution    JobExecution[]
+    DatasetItem  DatasetItem[]
+    JobExecution JobExecution[]
 
     @@index([projectId])
+    @@index([projectId, timestamp])
     @@index([sessionId])
     @@index([name])
     @@index([userId])
     @@index([id, userId])
-    @@index([externalId])
     @@index(timestamp)
-    @@index(release)
-    @@index(updatedAt)
     @@index(createdAt)
     @@index([tags(ops: ArrayOps)], type: Gin)
     @@map("traces")
@@ -292,27 +296,36 @@ model Observation {
     statusMessage       String?          @map("status_message")
     version             String?
     createdAt           DateTime         @default(now()) @map("created_at")
+    updatedAt           DateTime         @default(now()) @updatedAt @map("updated_at")
 
     // GENERATION ONLY
-    model               String?
-    internalModel       String?           @map("internal_model")
-    modelParameters     Json?
-    input               Json?
-    output              Json?
-    promptTokens        Int               @default(0) @map("prompt_tokens")
-    completionTokens    Int               @default(0) @map("completion_tokens")
-    totalTokens         Int               @default(0) @map("total_tokens")
-    unit                String?
-    inputCost           Decimal?          @map("input_cost")
-    outputCost          Decimal?          @map("output_cost")
-    totalCost           Decimal?          @map("total_cost")
-    completionStartTime DateTime?         @map("completion_start_time")
-    project             Project           @relation(fields: [projectId], references: [id], onDelete: Cascade)
-    derivedDatasetItems DatasetItem[]
-    datasetRunItems     DatasetRunItems[]
+    model           String?
+    internalModel   String? @map("internal_model")
+    internalModelId String? @map("internal_model_id")
 
-    promptId String? @map("prompt_id")
-    prompt   Prompt? @relation(fields: [promptId], onDelete: SetNull, references: [id])
+    modelParameters  Json?
+    input            Json?
+    output           Json?
+    promptTokens     Int     @default(0) @map("prompt_tokens")
+    completionTokens Int     @default(0) @map("completion_tokens")
+    totalTokens      Int     @default(0) @map("total_tokens")
+    unit             String?
+
+    // User provided cost at ingestion
+    inputCost  Decimal? @map("input_cost")
+    outputCost Decimal? @map("output_cost")
+    totalCost  Decimal? @map("total_cost")
+
+    // Calculated cost
+    calculatedInputCost  Decimal? @map("calculated_input_cost")
+    calculatedOutputCost Decimal? @map("calculated_output_cost")
+    calculatedTotalCost  Decimal? @map("calculated_total_cost")
+
+    completionStartTime DateTime?     @map("completion_start_time")
+    project             Project       @relation(fields: [projectId], references: [id], onDelete: Cascade)
+    derivedDatasetItems DatasetItem[]
+
+    promptId String? @map("prompt_id") // no fk constraint, prompt can be deleted
 
     @@unique([id, projectId])
     @@index([projectId, internalModel, startTime, unit])
@@ -324,9 +337,9 @@ model Observation {
     @@index(startTime)
     @@index(createdAt)
     @@index(projectId)
-    @@index(parentObservationId)
     @@index(model)
     @@index(internalModel)
+    @@index([projectId, promptId])
     @@index(promptId)
     @@index([projectId, startTime, type])
     @@map("observations")
@@ -403,46 +416,63 @@ model Score {
     comment       String?
     traceId       String         @map("trace_id")
     observationId String?        @map("observation_id")
+    configId      String?        @map("config_id")
+    stringValue   String?        @map("string_value")
+    createdAt     DateTime       @default(now()) @map("created_at")
+    updatedAt     DateTime       @default(now()) @updatedAt @map("updated_at")
+    dataType      ScoreDataType  @default(NUMERIC) @map("data_type")
     JobExecution  JobExecution[]
+    scoreConfig   ScoreConfig?   @relation(fields: [configId], references: [id], onDelete: SetNull)
 
     @@unique([id, projectId]) // used for upserts via prisma
     @@index(timestamp)
     @@index([value])
     @@index([projectId])
+    @@index([projectId, name])
     @@index([authorUserId])
+    @@index([configId])
     @@index([traceId], type: Hash)
     @@index([observationId], type: Hash)
     @@index([source])
+    @@index([createdAt])
     @@map("scores")
 }
 
 enum ScoreSource {
+    ANNOTATION
     API
-    REVIEW
     EVAL
 }
 
-enum PricingUnit {
-    PER_1000_TOKENS
-    PER_1000_CHARS
+model ScoreConfig {
+    id          String        @id @default(cuid())
+    createdAt   DateTime      @default(now()) @map("created_at")
+    updatedAt   DateTime      @default(now()) @updatedAt @map("updated_at")
+    projectId   String        @map("project_id")
+    project     Project       @relation(fields: [projectId], references: [id], onDelete: Cascade)
+    name        String
+    dataType    ScoreDataType @map("data_type")
+    isArchived  Boolean       @default(false) @map("is_archived")
+    minValue    Float?        @map("min_value")
+    maxValue    Float?        @map("max_value")
+    categories  Json?         @map("categories")
+    description String?
+    score       Score[]
+
+    @@unique([id, projectId]) // used for upserts via prisma
+    @@index([dataType])
+    @@index([isArchived])
+    @@index([projectId])
+    @@index([categories])
+    @@index([createdAt])
+    @@index([updatedAt])
+    @@map("score_configs")
 }
 
-enum TokenType {
-    PROMPT
-    COMPLETION
-    TOTAL
-}
-
-model Pricing {
-    id          String      @id @default(cuid())
-    modelName   String      @map("model_name")
-    pricingUnit PricingUnit @default(PER_1000_TOKENS) @map("pricing_unit")
-    price       Decimal
-    currency    String      @default("USD")
-    tokenType   TokenType   @map("token_type")
-
-    @@index(modelName)
-    @@map("pricings")
+enum ScoreDataType {
+    CATEGORICAL
+    NUMERIC
+    BOOLEAN
 }
 
 model CronJobs {
@@ -468,6 +498,8 @@ model Dataset {
 
     @@unique([projectId, name])
     @@index([projectId], type: Hash)
+    @@index([createdAt])
+    @@index([updatedAt])
     @@map("datasets")
 }
 
@@ -487,8 +519,11 @@ model DatasetItem {
     updatedAt           DateTime          @default(now()) @updatedAt @map("updated_at")
     datasetRunItems     DatasetRunItems[]
 
+    @@index([sourceTraceId], type: Hash)
     @@index([sourceObservationId], type: Hash)
     @@index([datasetId], type: Hash)
+    @@index([createdAt])
+    @@index([updatedAt])
     @@map("dataset_items")
 }
 
@@ -510,26 +545,28 @@ model DatasetRuns {
 
     @@unique([datasetId, name])
     @@index([datasetId], type: Hash)
+    @@index([createdAt])
+    @@index([updatedAt])
     @@map("dataset_runs")
 }
 
 model DatasetRunItems {
-    id            String       @id @default(cuid())
-    datasetRunId  String       @map("dataset_run_id")
-    datasetRun    DatasetRuns  @relation(fields: [datasetRunId], references: [id], onDelete: Cascade)
-    datasetItemId String       @map("dataset_item_id")
-    datasetItem   DatasetItem  @relation(fields: [datasetItemId], references: [id], onDelete: Cascade)
-    traceId       String       @map("trace_id")
-    trace         Trace        @relation(fields: [traceId], references: [id], onDelete: Cascade)
-    observationId String?      @map("observation_id")
-    observation   Observation? @relation(fields: [observationId], references: [id], onDelete: Cascade)
-    createdAt     DateTime     @default(now()) @map("created_at")
-    updatedAt     DateTime     @default(now()) @updatedAt @map("updated_at")
+    id            String      @id @default(cuid())
+    datasetRunId  String      @map("dataset_run_id")
+    datasetRun    DatasetRuns @relation(fields: [datasetRunId], references: [id], onDelete: Cascade)
+    datasetItemId String      @map("dataset_item_id")
+    datasetItem   DatasetItem @relation(fields: [datasetItemId], references: [id], onDelete: Cascade)
+    traceId       String      @map("trace_id")
+    observationId String?     @map("observation_id")
+    createdAt     DateTime    @default(now()) @map("created_at")
+    updatedAt     DateTime    @default(now()) @updatedAt @map("updated_at")
 
     @@index([datasetRunId], type: Hash)
     @@index([datasetItemId], type: Hash)
     @@index([observationId], type: Hash)
     @@index([traceId])
+    @@index([createdAt])
+    @@index([updatedAt])
     @@map("dataset_run_items")
 }
 
@@ -558,20 +595,21 @@ model Prompt {
 
     createdBy String @map("created_by")
 
-    prompt      Json
-    name        String
-    version     Int
-    type        String        @default("text")
-    isActive    Boolean?      @map("is_active") // Deprecated. To be removed once 'production' labels work as expected.
-    config      Json          @default("{}")
-    tags        String[]      @default([])
-    labels      String[]      @default([])
-    Observation Observation[]
+    prompt   Json
+    name     String
+    version  Int
+    type     String   @default("text")
+    isActive Boolean? @map("is_active") // Deprecated. To be removed once 'production' labels work as expected.
+    config   Json     @default("{}")
+    tags     String[] @default([])
+    labels   String[] @default([])
 
     @@unique([projectId, name, version])
     @@index([projectId, name, version])
     @@index([projectId, id])
     @@index([projectId])
+    @@index([createdAt])
+    @@index([updatedAt])
     @@index([tags(ops: ArrayOps)], type: Gin)
     @@map("prompts")
 }
@@ -619,6 +657,7 @@ model AuditLog {
 
     @@index([projectId])
     @@index([createdAt])
+    @@index([updatedAt])
     @@map("audit_logs")
 }
 
@@ -634,6 +673,7 @@ model EvalTemplate {
     version          Int
     prompt           String
     model            String
+    provider         String
     modelParams      Json               @map("model_params")
     vars             String[]           @default([])
     outputSchema     Json               @map("output_schema")
@@ -712,6 +752,11 @@ model JobExecution {
     @@index([projectId, status])
     @@index([projectId, id])
     @@index([projectId])
+    @@index([jobConfigurationId])
+    @@index([jobOutputScoreId])
+    @@index([jobInputTraceId])
+    @@index([createdAt])
+    @@index([updatedAt])
     @@map("job_executions")
 }
 
@@ -742,3 +787,27 @@ model PosthogIntegration {
     @@index([projectId])
     @@map("posthog_integrations")
 }
+
+model BatchExport {
+    id        String   @id @default(cuid())
+    createdAt DateTime @default(now()) @map("created_at")
+    updatedAt DateTime @default(now()) @updatedAt @map("updated_at")
+
+    projectId String  @map("project_id")
+    project   Project @relation(fields: [projectId], references: [id], onDelete: Cascade)
+    userId    String  @map("user_id")
+
+    finishedAt DateTime? @map("finished_at")
+    expiresAt  DateTime? @map("expires_at")
+
+    name   String
+    status String
+    query  Json
+    format String
+    url    String?
+    log    String?
+
+    @@index([projectId, userId])
+    @@index([status])
+    @@map("batch_exports")
+}

packages/shared/prisma/seed.ts

@@ -4,6 +4,7 @@ import {
   type Prisma,
   ObservationType,
   ScoreSource,
+  ScoreDataType,
 } from "../src/index";
 import { hash } from "bcryptjs";
 import { parseArgs } from "node:util";
@@ -11,7 +12,8 @@ import { parseArgs } from "node:util";
 import { chunk } from "lodash";
 import { v4 } from "uuid";
 import { ModelUsageUnit } from "../src";
-import { getDisplaySecretKey, hashSecretKey } from "../src/server/auth";
+import { getDisplaySecretKey, hashSecretKey } from "../src/server";
+import { encrypt } from "../src/encryption";
 
 const LOAD_TRACE_VOLUME = 10_000;
 
@@ -38,6 +40,7 @@ async function main() {
       name: "Demo User",
       email: "demo@langfuse.com",
       password: await hash("password", 12),
+      image: "https://static.langfuse.com/langfuse-dev%2Fexample-avatar.png",
     },
   });
 
@@ -143,6 +146,11 @@ async function main() {
       });
     }
 
+    const configIdsAndNames = await generateConfigsForProject([
+      project1,
+      project2,
+    ]);
+
     const promptIds = await generatePromptsForProject([project1, project2]);
 
     const envTags = [null, "development", "staging", "production"];
@@ -156,7 +164,8 @@ async function main() {
       colorTags,
       project1,
       project2,
-      promptIds
+      promptIds,
+      configIdsAndNames
     );
 
     console.log(
@@ -165,6 +174,25 @@ async function main() {
 
     await uploadObjects(traces, observations, scores, sessions, events);
 
+    // If openai key is in environment, add it to the projects LLM API keys
+    const OPENAI_API_KEY = process.env.OPENAI_API_KEY;
+
+    if (OPENAI_API_KEY) {
+      await prisma.llmApiKeys.create({
+        data: {
+          projectId: project1.id,
+          secretKey: encrypt(OPENAI_API_KEY),
+          displaySecretKey: getDisplaySecretKey(OPENAI_API_KEY),
+          provider: "openai",
+          adapter: "openai",
+        },
+      });
+    } else {
+      console.warn(
+        "No OPENAI_API_KEY found in environment. Skipping seeding LLM API key."
+      );
+    }
+
     // add eval objects
     const evalTemplate = await prisma.evalTemplate.upsert({
       where: {
@@ -182,6 +210,7 @@ async function main() {
           "Please evaluate the toxicity of the following text {{input}} {{output}}",
         model: "gpt-3.5-turbo",
         vars: ["input", "output"],
+        provider: "openai",
         outputSchema: {
           score: "provide a score between 0 and 1",
           reasoning: "one sentence reasoning for the score",
@@ -348,11 +377,10 @@ async function uploadObjects(
   });
 
   for (let i = 0; i < promises.length; i++) {
-    process.stdout.clearLine(0);
-    process.stdout.cursorTo(0);
-    process.stdout.write(
-      `Seeding of Sessions ${(i / promises.length) * 100}% complete`
-    );
+    if (i + 1 >= promises.length || i % Math.ceil(promises.length / 10) === 0)
+      console.log(
+        `Seeding of Sessions ${((i + 1) / promises.length) * 100}% complete`
+      );
     await promises[i];
   }
 
@@ -366,11 +394,10 @@ async function uploadObjects(
     );
   });
   for (let i = 0; i < promises.length; i++) {
-    process.stdout.clearLine(0);
-    process.stdout.cursorTo(0);
-    process.stdout.write(
-      `Seeding of Traces ${(i / promises.length) * 100}% complete`
-    );
+    if (i + 1 >= promises.length || i % Math.ceil(promises.length / 10) === 0)
+      console.log(
+        `Seeding of Traces ${((i + 1) / promises.length) * 100}% complete`
+      );
     await promises[i];
   }
 
@@ -384,11 +411,10 @@ async function uploadObjects(
   });
 
   for (let i = 0; i < promises.length; i++) {
-    process.stdout.clearLine(0);
-    process.stdout.cursorTo(0);
-    process.stdout.write(
-      `Seeding of Observations ${(i / promises.length) * 100}% complete`
-    );
+    if (i + 1 >= promises.length || i % Math.ceil(promises.length / 10) === 0)
+      console.log(
+        `Seeding of Observations ${((i + 1) / promises.length) * 100}% complete`
+      );
     await promises[i];
   }
 
@@ -402,11 +428,10 @@ async function uploadObjects(
   });
 
   for (let i = 0; i < promises.length; i++) {
-    process.stdout.clearLine(0);
-    process.stdout.cursorTo(0);
-    process.stdout.write(
-      `Seeding of Events ${(i / promises.length) * 100}% complete`
-    );
+    if (i + 1 >= promises.length || i % Math.ceil(promises.length / 10) === 0)
+      console.log(
+        `Seeding of Events ${((i + 1) / promises.length) * 100}% complete`
+      );
     await promises[i];
   }
 
@@ -419,11 +444,10 @@ async function uploadObjects(
     );
   });
   for (let i = 0; i < promises.length; i++) {
-    process.stdout.clearLine(0);
-    process.stdout.cursorTo(0);
-    process.stdout.write(
-      `Seeding of Scores ${(i / promises.length) * 100}% complete`
-    );
+    if (i + 1 >= promises.length || i % Math.ceil(promises.length / 10) === 0)
+      console.log(
+        `Seeding of Scores ${((i + 1) / promises.length) * 100}% complete`
+      );
     await promises[i];
   }
 }
@@ -434,13 +458,15 @@ function createObjects(
   colorTags: (string | null)[],
   project1: Project,
   project2: Project,
-  promptIds: Map<string, string[]>
+  promptIds: Map<string, string[]>,
+  configIdsAndNames: Map<string, { name: string; id: string }[]>
 ) {
   const traces: Prisma.TraceCreateManyInput[] = [];
   const observations: Prisma.ObservationCreateManyInput[] = [];
   const scores: Prisma.ScoreCreateManyInput[] = [];
   const sessions: Prisma.TraceSessionCreateManyInput[] = [];
   const events: Prisma.ObservationCreateManyInput[] = [];
+  const configs: Prisma.ScoreConfigCreateManyInput[] = [];
 
   for (let i = 0; i < traceVolume; i++) {
     // print progress to console with a progress bar that refreshes every 10 iterations
@@ -492,17 +518,28 @@ function createObjects(
 
     traces.push(trace);
 
+    const configArray = configIdsAndNames.get(projectId) ?? [];
+    const randomIndex = Math.floor(Math.random() * 3);
+    const config =
+      configArray.length >= randomIndex - 1 && configArray[randomIndex];
+    const { name: annotationScoreName, id: configId } = config || {
+      name: "manual-score",
+      id: undefined,
+    };
+
     const traceScores = [
       ...(Math.random() > 0.5
         ? [
             {
               traceId: trace.id,
-              name: "manual-score",
+              name: annotationScoreName,
               value: Math.floor(Math.random() * 3) - 1,
               timestamp: traceTs,
-              source: ScoreSource.REVIEW,
+              source: ScoreSource.ANNOTATION,
               projectId,
               authorUserId: `user-${i}`,
+              dataType: ScoreDataType.NUMERIC,
+              ...(configId ? { configId } : {}),
             },
           ]
         : []),
@@ -515,6 +552,7 @@ function createObjects(
               timestamp: traceTs,
               source: ScoreSource.API,
               projectId,
+              dataType: ScoreDataType.NUMERIC,
             },
           ]
         : []),
@@ -753,6 +791,7 @@ function createObjects(
     traces,
     observations,
     scores,
+    configs,
     sessions: uniqueSessions,
     events,
   };
@@ -935,3 +974,80 @@ async function generatePrompts(project: Project) {
   }
   return promptIds;
 }
+
+async function generateConfigsForProject(projects: Project[]) {
+  const projectIdsToConfigs: Map<string, { name: string; id: string }[]> =
+    new Map();
+
+  await Promise.all(
+    projects.map(async (project) => {
+      const configNameAndId = await generateConfigs(project);
+      projectIdsToConfigs.set(project.id, configNameAndId);
+    })
+  );
+  return projectIdsToConfigs;
+}
+
+async function generateConfigs(project: Project) {
+  const configNameAndId: { name: string; id: string }[] = [];
+
+  const configs = [
+    {
+      id: `config-${v4()}`,
+      name: "manual-score",
+      dataType: ScoreDataType.NUMERIC,
+      projectId: project.id,
+      isArchived: false,
+    },
+    {
+      id: `config-${v4()}`,
+      projectId: project.id,
+      name: "Accuracy",
+      dataType: ScoreDataType.CATEGORICAL,
+      categories: [
+        { label: "Incorrect", value: 0 },
+        { label: "Partially Correct", value: 1 },
+        { label: "Correct", value: 2 },
+      ],
+      isArchived: false,
+    },
+    {
+      id: `config-${v4()}`,
+      projectId: project.id,
+      name: "Toxicity",
+      dataType: ScoreDataType.BOOLEAN,
+      categories: [
+        { label: "True", value: 1 },
+        { label: "False", value: 0 },
+      ],
+      description:
+        "Used to indicate if text was harmful or offensive in nature.",
+      isArchived: false,
+    },
+  ];
+
+  for (const config of configs) {
+    await prisma.scoreConfig.upsert({
+      where: {
+        id_projectId: {
+          projectId: config.projectId,
+          id: config.id,
+        },
+      },
+      create: {
+        id: config.id,
+        projectId: config.projectId,
+        name: config.name,
+        dataType: config.dataType,
+        categories: config.categories,
+        isArchived: config.isArchived,
+      },
+      update: {
+        id: config.id,
+      },
+    });
+    configNameAndId.push({ name: config.name, id: config.id });
+  }
+
+  return configNameAndId;
+}

packages/shared/scripts/cleanup.sql

@@ -0,0 +1,7 @@
+DO $$
+BEGIN
+ IF EXISTS (SELECT 1 FROM information_schema.tables WHERE table_name = '_prisma_migrations') THEN
+  DELETE FROM _prisma_migrations
+  WHERE migration_name IN ('20240606090858_pricings_add_latest_gemini_models', '20240530212419_model_price_anthropic_via_google_vertex', '20240604133340_backfill_manual_scores');
+ END IF;
+END $$;

packages/shared/src/features/annotation/types.ts

@@ -0,0 +1,15 @@
+import z from "zod";
+import { ScoreConfig } from "../../db";
+
+const configCategory = z.object({
+  label: z.string().min(1),
+  value: z.number(),
+});
+
+export const categoriesList = z.array(configCategory);
+
+export type ConfigCategory = z.infer<typeof configCategory>;
+
+export type CastedConfig = Omit<ScoreConfig, "categories"> & {
+  categories: ConfigCategory[] | null;
+};

packages/shared/src/features/batchExport/types.ts

@@ -0,0 +1,73 @@
+import z from "zod";
+
+import { BatchExport } from "@prisma/client";
+
+import { singleFilter } from "../../interfaces/filters";
+import { orderBy } from "../../interfaces/orderBy";
+
+export enum BatchExportStatus {
+  QUEUED = "QUEUED",
+  PROCESSING = "PROCESSING",
+  COMPLETED = "COMPLETED",
+  FAILED = "FAILED",
+}
+
+export enum BatchExportFileFormat {
+  JSON = "JSON",
+  CSV = "CSV",
+}
+
+export enum BatchExportTableName {
+  Sessions = "sessions",
+}
+
+export const exportOptions: Record<
+  BatchExportFileFormat,
+  {
+    label: string;
+    extension: string;
+    fileType: string;
+  }
+> = {
+  CSV: { label: "CSV", extension: "csv", fileType: "text/csv" },
+  JSON: { label: "JSON", extension: "json", fileType: "application/json" },
+} as const;
+
+export const BatchExportQuerySchema = z.object({
+  tableName: z.nativeEnum(BatchExportTableName),
+  filter: z.array(singleFilter).nullable(),
+  orderBy,
+  limit: z.number().optional(),
+  page: z.number().optional(),
+});
+
+export type BatchExportQueryType = z.infer<typeof BatchExportQuerySchema>;
+
+export const CreateBatchExportSchema = z.object({
+  projectId: z.string(),
+  name: z.string(),
+  query: BatchExportQuerySchema,
+  format: z.nativeEnum(BatchExportFileFormat),
+});
+
+export const BatchExportSchema = z.object({
+  id: z.string(),
+  createdAt: z.date(),
+  updatedAt: z.date(),
+  projectId: z.string(),
+  userId: z.string(),
+  finishedAt: z.date().nullable(),
+  expiresAt: z.date().nullable(),
+  name: z.string(),
+  status: z.nativeEnum(BatchExportStatus),
+  query: BatchExportQuerySchema,
+  format: z.nativeEnum(BatchExportFileFormat),
+  url: z.string().nullable(),
+  log: z.string().nullable(),
+});
+
+// Ensure that zod type matches the Prisma type
+export type BatchExportType =
+  z.infer<typeof BatchExportSchema> extends BatchExport
+    ? z.infer<typeof BatchExportSchema>
+    : never;

packages/shared/src/features/evals/types.ts

@@ -1,5 +1,5 @@
 import z from "zod";
-import { ModelProvider, UIModelParams } from "../..";
+import { LLMAdapter } from "../..";
 
 export const langfuseObjects = [
   "trace",
@@ -77,38 +77,15 @@ export const availableEvalVariables = [
   },
 ];
 
-export const evalLLMModels: UIModelParams[] = [
-  {
-    provider: { value: ModelProvider.OpenAI, enabled: true },
-    model: { value: "gpt-3.5-turbo", enabled: true },
-    temperature: { value: 1, enabled: true },
-    maxTemperature: { value: 2, enabled: true },
-    max_tokens: { value: 256, enabled: true },
-    top_p: { value: 1, enabled: true },
-  },
-  {
-    provider: { value: ModelProvider.OpenAI, enabled: true },
-    model: { value: "gpt-4-turbo-preview", enabled: true },
-    temperature: { value: 1, enabled: true },
-    maxTemperature: { value: 2, enabled: true },
-    max_tokens: { value: 256, enabled: true },
-    top_p: { value: 1, enabled: true },
-  },
-  {
-    provider: { value: ModelProvider.OpenAI, enabled: true },
-    model: { value: "gpt-4o", enabled: true },
-    temperature: { value: 1, enabled: true },
-    maxTemperature: { value: 2, enabled: true },
-    max_tokens: { value: 256, enabled: true },
-    top_p: { value: 1, enabled: true },
-  },
-] as const;
-
-export const EvalModelNames = z.enum([
-  "gpt-3.5-turbo",
-  "gpt-4-turbo-preview",
-  "gpt-4o",
-]);
+export const evalModelsByAdapter: Record<LLMAdapter, string[]> = {
+  [LLMAdapter.OpenAI]: [
+    "gpt-4o",
+    "gpt-4-turbo-preview",
+    "gpt-3.5-turbo",
+  ] as const,
+  [LLMAdapter.Anthropic]: [] as const,
+  [LLMAdapter.Azure]: [] as const,
+};
 
 export const OutputSchema = z.object({
   reasoning: z.string(),

packages/shared/src/features/ingestion/types.ts

@@ -1,9 +1,9 @@
 import lodash from "lodash";
 import { z } from "zod";
 
-import { ModelUsageUnit } from "@langfuse/shared";
-import { NonEmptyString, jsonSchema } from "@/src/utils/zod";
-import { ObservationLevel } from "@langfuse/shared";
+import { NonEmptyString, jsonSchema } from "../../utils/zod";
+import { ModelUsageUnit } from "../../constants";
+import { ObservationLevel } from "@prisma/client";
 
 export const Usage = z.object({
   input: z.number().int().nullish(),
@@ -109,7 +109,7 @@ export const CreateGenerationBody = CreateSpanBody.extend({
       z.string(),
       z
         .union([z.string(), z.number(), z.boolean(), z.array(z.string())])
-        .nullish(),
+        .nullish()
     )
     .nullish(),
   usage: usage,
@@ -131,7 +131,7 @@ export const UpdateGenerationBody = UpdateSpanBody.extend({
       z.string(),
       z
         .union([z.string(), z.number(), z.boolean(), z.array(z.string())])
-        .nullish(),
+        .nullish()
     )
     .nullish(),
   usage: usage,
@@ -195,7 +195,7 @@ export const LegacyGenerationsCreateSchema = z.object({
   modelParameters: z
     .record(
       z.string(),
-      z.union([z.string(), z.number(), z.boolean()]).nullish(),
+      z.union([z.string(), z.number(), z.boolean()]).nullish()
     )
     .nullish(),
   prompt: jsonSchema.nullish(),
@@ -219,7 +219,7 @@ export const LegacyGenerationPatchSchema = z.object({
   modelParameters: z
     .record(
       z.string(),
-      z.union([z.string(), z.number(), z.boolean()]).nullish(),
+      z.union([z.string(), z.number(), z.boolean()]).nullish()
     )
     .nullish(),
   prompt: jsonSchema.nullish(),
@@ -243,7 +243,7 @@ export const LegacyObservationBody = z.object({
   modelParameters: z
     .record(
       z.string(),
-      z.union([z.string(), z.number(), z.boolean()]).nullish(),
+      z.union([z.string(), z.number(), z.boolean()]).nullish()
     )
     .nullish(),
   input: jsonSchema.nullish(),

packages/shared/src/filterToPrisma.ts

@@ -1,8 +1,5 @@
 import { Prisma } from "@prisma/client";
-import {
-  ColumnDefinition,
-  type TableNames as TableName,
-} from "./interfaces/tableDefinition";
+import { ColumnDefinition, type TableNames } from "./tableDefinitions";
 import { FilterState } from "./types";
 import { filterOperators } from "./interfaces/filters";
 
@@ -27,7 +24,7 @@ const arrayOperatorReplacements = {
 export function tableColumnsToSqlFilterAndPrefix(
   filters: FilterState,
   tableColumns: ColumnDefinition[],
-  table: TableName
+  table: TableNames
 ): Prisma.Sql {
   const sql = tableColumnsToSqlFilter(filters, tableColumns, table);
   if (sql === Prisma.empty) {
@@ -43,7 +40,7 @@ export function tableColumnsToSqlFilterAndPrefix(
 export function tableColumnsToSqlFilter(
   filters: FilterState,
   tableColumns: ColumnDefinition[],
-  table: TableName
+  table: TableNames
 ): Prisma.Sql {
   const internalFilters = filters.map((filter) => {
     // Get column definition to map column to internal name, e.g. "t.id"
@@ -153,7 +150,7 @@ export function tableColumnsToSqlFilter(
 
 const castValueToPostgresTypes = (
   column: ColumnDefinition,
-  table: TableName
+  table: TableNames
 ) => {
   return column.name === "type" &&
     (table === "observations" ||

packages/shared/src/index.ts

@@ -1,14 +1,18 @@
 export * from "./constants";
+export * from "./queries";
 export * from "./queues";
-export * from "./interfaces/exportTypes";
 export * from "./interfaces/filters";
 export * from "./interfaces/orderBy";
-export * from "./interfaces/tableDefinition";
+export * from "./tableDefinitions";
 export * from "./types";
 export * from "./filterToPrisma";
+export * from "./orderByToPrisma";
 export * from "./tracesTable";
 export * from "./server/auth";
 export * from "./observationsTable";
+export * from "./features/ingestion/types";
+export * from "./utils/zod";
+export * from "./utils/json";
 export { env } from "./env";
 
 // llm api
@@ -17,6 +21,10 @@ export * from "./server/llm/fetchLLMCompletion";
 
 // evals
 export * from "./features/evals/types";
+export * from "./features/batchExport/types";
+
+// annotation
+export * from "./features/annotation/types";
 
 // export db types only
 export * from "@prisma/client";

packages/shared/src/interfaces/exportTypes.ts

@@ -1,19 +0,0 @@
-export const exportFileFormats = ["CSV", "JSON", "OPENAI-JSONL"] as const;
-export type ExportFileFormats = (typeof exportFileFormats)[number];
-
-export const exportOptions: Record<
-  ExportFileFormats,
-  {
-    label: string;
-    extension: string;
-    fileType: string;
-  }
-> = {
-  CSV: { label: "CSV", extension: "csv", fileType: "text/csv" },
-  JSON: { label: "JSON", extension: "json", fileType: "application/json" },
-  "OPENAI-JSONL": {
-    label: "OpenAI JSONL (fine-tuning)",
-    extension: "jsonl",
-    fileType: "application/json",
-  },
-} as const;

packages/shared/src/interfaces/orderBy.ts

@@ -6,3 +6,5 @@ export const orderBy = z
     order: z.enum(["ASC", "DESC"]),
   })
   .nullable();
+
+export type OrderByState = z.infer<typeof orderBy>;

packages/shared/src/observationsTable.ts

@@ -2,7 +2,7 @@ import { ObservationLevel } from "@prisma/client";
 import {
   type OptionsDefinition,
   type ColumnDefinition,
-} from "./interfaces/tableDefinition";
+} from "./tableDefinitions/types";
 
 // to be used server side
 export const observationsTableCols: ColumnDefinition[] = [