phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-28 02:11:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'main' into out-of-token-budget-fix

Browse source
This commit is contained in:
Ilya Kolchinsky 2025-05-14 12:46:24 +02:00 • committed by GitHub
commit 85ef55391d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
61 changed files with 1322 additions and 1598 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama_stack/providers/remote/datasetio/huggingface/huggingface.py
View file

@ -12,8 +12,8 @@ from llama_stack.apis.common.responses import PaginatedResponse
from llama_stack.apis.datasetio import DatasetIO
from llama_stack.apis.datasets import Dataset
from llama_stack.providers.datatypes import DatasetsProtocolPrivate
from llama_stack.providers.utils.datasetio.pagination import paginate_records
from llama_stack.providers.utils.kvstore import kvstore_impl
from llama_stack.providers.utils.pagination import paginate_records
from .config import HuggingfaceDatasetIOConfig

97
llama_stack/providers/remote/inference/ollama/ollama.py
View file

@ -61,6 +61,7 @@ from llama_stack.providers.utils.inference.openai_compat import (
OpenAICompatCompletionChoice,
OpenAICompatCompletionResponse,
get_sampling_options,
prepare_openai_completion_params,
process_chat_completion_response,
process_chat_completion_stream_response,
process_completion_response,
@ -395,29 +396,25 @@ class OllamaInferenceAdapter(
raise ValueError("Ollama does not support non-string prompts for completion")
model_obj = await self._get_model(model)
params = {
k: v
for k, v in {
"model": model_obj.provider_resource_id,
"prompt": prompt,
"best_of": best_of,
"echo": echo,
"frequency_penalty": frequency_penalty,
"logit_bias": logit_bias,
"logprobs": logprobs,
"max_tokens": max_tokens,
"n": n,
"presence_penalty": presence_penalty,
"seed": seed,
"stop": stop,
"stream": stream,
"stream_options": stream_options,
"temperature": temperature,
"top_p": top_p,
"user": user,
}.items()
if v is not None
}
params = await prepare_openai_completion_params(
model=model_obj.provider_resource_id,
prompt=prompt,
best_of=best_of,
echo=echo,
frequency_penalty=frequency_penalty,
logit_bias=logit_bias,
logprobs=logprobs,
max_tokens=max_tokens,
n=n,
presence_penalty=presence_penalty,
seed=seed,
stop=stop,
stream=stream,
stream_options=stream_options,
temperature=temperature,
top_p=top_p,
user=user,
)
return await self.openai_client.completions.create(**params) # type: ignore
async def openai_chat_completion(
@ -447,35 +444,31 @@ class OllamaInferenceAdapter(
user: str | None = None,
) -> OpenAIChatCompletion | AsyncIterator[OpenAIChatCompletionChunk]:
model_obj = await self._get_model(model)
params = {
k: v
for k, v in {
"model": model_obj.provider_resource_id,
"messages": messages,
"frequency_penalty": frequency_penalty,
"function_call": function_call,
"functions": functions,
"logit_bias": logit_bias,
"logprobs": logprobs,
"max_completion_tokens": max_completion_tokens,
"max_tokens": max_tokens,
"n": n,
"parallel_tool_calls": parallel_tool_calls,
"presence_penalty": presence_penalty,
"response_format": response_format,
"seed": seed,
"stop": stop,
"stream": stream,
"stream_options": stream_options,
"temperature": temperature,
"tool_choice": tool_choice,
"tools": tools,
"top_logprobs": top_logprobs,
"top_p": top_p,
"user": user,
}.items()
if v is not None
}
params = await prepare_openai_completion_params(
model=model_obj.provider_resource_id,
messages=messages,
frequency_penalty=frequency_penalty,
function_call=function_call,
functions=functions,
logit_bias=logit_bias,
logprobs=logprobs,
max_completion_tokens=max_completion_tokens,
max_tokens=max_tokens,
n=n,
parallel_tool_calls=parallel_tool_calls,
presence_penalty=presence_penalty,
response_format=response_format,
seed=seed,
stop=stop,
stream=stream,
stream_options=stream_options,
temperature=temperature,
tool_choice=tool_choice,
tools=tools,
top_logprobs=top_logprobs,
top_p=top_p,
user=user,
)
return await self.openai_client.chat.completions.create(**params) # type: ignore
async def batch_completion(

3
llama_stack/providers/remote/vector_io/chroma/chroma.py
View file

@ -26,8 +26,7 @@ from .config import ChromaVectorIOConfig as RemoteChromaVectorIOConfig
log = logging.getLogger(__name__)
ChromaClientType = chromadb.AsyncHttpClient | chromadb.PersistentClient
ChromaClientType = chromadb.api.AsyncClientAPI | chromadb.api.ClientAPI
# this is a helper to allow us to use async and non-async chroma clients interchangeably