Files
zeekay 925e62aae4 release(v1.7.43): Vulkan DSL MoE + demand-sized KV cache; hanzo-ml 0.11.55
- Demand-sized unified-memory KV cache: size to max_seq_len*max_batch_size
  (context*concurrency) bounded by the post-model/OS-reserve ceiling, instead
  of grabbing all free RAM. Fixes the ROCm/APU thrash where an ~88 GB grab on a
  128 GB unified box never returned; 30B-A3B now 88 GB -> 24 GB, elastic with
  concurrency. (hanzo-engine/src/paged_attention/mod.rs)
- Vulkan paged-attention wired (is_vulkan branch, opt-in via --paged-attn on):
  runs and eliminates the naive-Sdpa layout-copy churn, but the v1 scalar kernel
  regresses decode on this APU, so default OFF until v2 + MoE-glue fusion land.
  i64/i32 paged-attn index metadata now binds directly (backend coerces to u32).
- hanzo-quant: Vulkan/Wgpu storage arms in the distributed pipeline.
- Drop the temporary [patch.crates-io] hanzo-ml redirect; pin hanzo-ml 0.11.55.
- Manifest hygiene: version on path deps (enso, ffi, server-core, zen3-serving).
2026-07-14 17:26:58 -07:00

34 lines
972 B
TOML

[package]
name = "zen3-serving"
version = "0.1.0"
edition = "2021"
license = "MIT"
description = "Standalone CPU harness exercising the engine's zen3-ASR + zen3-TTS speech pipelines on real weights."
[[bin]]
name = "zen3-serving"
path = "src/main.rs"
[dependencies]
hanzo-engine = { version = "1.7.41", path = "../../hanzo-engine" }
hanzo-ml = { path = "../../../ml/hanzo-ml" }
hanzo-nn = { path = "../../../ml/hanzo-nn" }
hanzo-quant = { version = "1.7.41", path = "../../hanzo-quant" }
hanzo-audio = { version = "1.7.41", path = "../../hanzo-audio" }
anyhow = "1"
clap = { version = "4", features = ["derive"] }
tokenizers = { version = "0.21", default-features = false, features = ["onig"] }
serde_json = "1"
ahash = "0.8"
hound = "3.5"
[features]
cuda = ["hanzo-engine/cuda", "hanzo-ml/cuda", "hanzo-nn/cuda", "hanzo-quant/cuda"]
metal = ["hanzo-engine/metal", "hanzo-ml/metal", "hanzo-nn/metal", "hanzo-quant/metal"]
[profile.release]
opt-level = 3
[workspace]