mirror of
https://github.com/zenlm/engine.git
synced 2026-10-30 15:00:05 +00:00
c3be7b6be4
* Add model * Update loader * Fix llama config usage * Docs * Fix config no_rope_layers * Fix tie_word_embeddings default * Add chat template * Embed the chat templates * Fix embedding template * enable_thinking default true * Update examples * XML tools for smollm3 * Add smollm3 docs * Fix openai examples * Clippy --------- Co-authored-by: Eric Buehler <[email protected]>
48 lines
1012 B
Python
48 lines
1012 B
Python
from mistralrs import Runner, Which, ChatCompletionRequest, Architecture
|
|
from json import dumps
|
|
|
|
runner = Runner(
|
|
which=Which.Plain(
|
|
model_id="microsoft/Phi-3.5-mini-instruct",
|
|
),
|
|
num_device_layers=["500"],
|
|
)
|
|
|
|
# see lark_llg.py for a better way of dealing with JSON and Lark together
|
|
|
|
json_lark = r"""
|
|
start: object # we only want objects
|
|
|
|
value: object
|
|
| array
|
|
| STRING
|
|
| NUMBER
|
|
| "true"
|
|
| "false"
|
|
| "null"
|
|
|
|
object: "{" [pair ("," pair)*] "}"
|
|
pair: STRING ":" value
|
|
|
|
array: "[" [value ("," value)*] "]"
|
|
|
|
STRING: /"(\\.|[^"\\])*"/
|
|
NUMBER: /-?(0|[1-9][0-9]*)(\.[0-9]+)?([eE][+-]?[0-9]+)?/
|
|
|
|
%import common.WS
|
|
%ignore WS
|
|
"""
|
|
|
|
res = runner.send_chat_completion_request(
|
|
ChatCompletionRequest(
|
|
model="default",
|
|
messages=[{"role": "user", "content": "Give me a sample address."}],
|
|
max_tokens=30,
|
|
temperature=0.1,
|
|
grammar_type="lark",
|
|
grammar=json_lark,
|
|
)
|
|
)
|
|
print(res.choices[0].message.content)
|
|
print(res.usage)
|