mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-06-28 02:53:30 +00:00
04de2f84e9
Each model known to the system has two identifiers: - the `provider_resource_id` (what the provider calls it) -- e.g., `accounts/fireworks/models/llama-v3p1-8b-instruct` - the `identifier` (`model_id`) under which it is registered and gets routed to the appropriate provider. We have so far used the HuggingFace repo alias as the standardized identifier you can use to refer to the model. So in the above example, we'd use `meta-llama/Llama-3.1-8B-Instruct` as the name under which it gets registered. This makes it convenient for users to refer to these models across providers. However, we forgot to register the _actual_ provider model ID also. You should be able to route via `provider_resource_id` also, of course. This change fixes this (somewhat grave) omission. *Note*: this change is additive -- more aliases work now compared to before. ## Test Plan Run the following for distro=(ollama fireworks together) ``` LLAMA_STACK_CONFIG=$distro \ pytest -s -v tests/client-sdk/inference/test_text_inference.py \ --inference-model=meta-llama/Llama-3.1-8B-Instruct --vision-inference-model="" ```
127 lines
4.3 KiB
Python
127 lines
4.3 KiB
Python
# Copyright (c) Meta Platforms, Inc. and affiliates.
|
|
# All rights reserved.
|
|
#
|
|
# This source code is licensed under the terms described in the LICENSE file in
|
|
# the root directory of this source tree.
|
|
|
|
from llama_stack.apis.models.models import ModelType
|
|
from llama_stack.distribution.datatypes import ModelInput, Provider
|
|
from llama_stack.providers.inline.inference.sentence_transformers import (
|
|
SentenceTransformersInferenceConfig,
|
|
)
|
|
from llama_stack.providers.inline.inference.vllm import VLLMConfig
|
|
from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
|
|
from llama_stack.templates.template import (
|
|
DistributionTemplate,
|
|
RunConfigSettings,
|
|
ToolGroupInput,
|
|
)
|
|
|
|
|
|
def get_distribution_template() -> DistributionTemplate:
|
|
providers = {
|
|
"inference": ["inline::vllm", "inline::sentence-transformers"],
|
|
"vector_io": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
|
|
"safety": ["inline::llama-guard"],
|
|
"agents": ["inline::meta-reference"],
|
|
"telemetry": ["inline::meta-reference"],
|
|
"eval": ["inline::meta-reference"],
|
|
"datasetio": ["remote::huggingface", "inline::localfs"],
|
|
"scoring": ["inline::basic", "inline::llm-as-judge", "inline::braintrust"],
|
|
"tool_runtime": [
|
|
"remote::brave-search",
|
|
"remote::tavily-search",
|
|
"inline::code-interpreter",
|
|
"inline::rag-runtime",
|
|
"remote::model-context-protocol",
|
|
],
|
|
}
|
|
|
|
name = "vllm-gpu"
|
|
inference_provider = Provider(
|
|
provider_id="vllm",
|
|
provider_type="inline::vllm",
|
|
config=VLLMConfig.sample_run_config(),
|
|
)
|
|
vector_io_provider = Provider(
|
|
provider_id="faiss",
|
|
provider_type="inline::faiss",
|
|
config=FaissVectorIOConfig.sample_run_config(f"distributions/{name}"),
|
|
)
|
|
embedding_provider = Provider(
|
|
provider_id="sentence-transformers",
|
|
provider_type="inline::sentence-transformers",
|
|
config=SentenceTransformersInferenceConfig.sample_run_config(),
|
|
)
|
|
|
|
inference_model = ModelInput(
|
|
model_id="${env.INFERENCE_MODEL}",
|
|
provider_id="vllm",
|
|
)
|
|
embedding_model = ModelInput(
|
|
model_id="all-MiniLM-L6-v2",
|
|
provider_id="sentence-transformers",
|
|
model_type=ModelType.embedding,
|
|
metadata={
|
|
"embedding_dimension": 384,
|
|
},
|
|
)
|
|
default_tool_groups = [
|
|
ToolGroupInput(
|
|
toolgroup_id="builtin::websearch",
|
|
provider_id="tavily-search",
|
|
),
|
|
ToolGroupInput(
|
|
toolgroup_id="builtin::rag",
|
|
provider_id="rag-runtime",
|
|
),
|
|
ToolGroupInput(
|
|
toolgroup_id="builtin::code_interpreter",
|
|
provider_id="code-interpreter",
|
|
),
|
|
]
|
|
|
|
return DistributionTemplate(
|
|
name=name,
|
|
distro_type="self_hosted",
|
|
description="Use a built-in vLLM engine for running LLM inference",
|
|
container_image=None,
|
|
template_path=None,
|
|
providers=providers,
|
|
run_configs={
|
|
"run.yaml": RunConfigSettings(
|
|
provider_overrides={
|
|
"inference": [inference_provider, embedding_provider],
|
|
"vector_io": [vector_io_provider],
|
|
},
|
|
default_models=[inference_model, embedding_model],
|
|
default_tool_groups=default_tool_groups,
|
|
),
|
|
},
|
|
run_config_env_vars={
|
|
"LLAMA_STACK_PORT": (
|
|
"5001",
|
|
"Port for the Llama Stack distribution server",
|
|
),
|
|
"INFERENCE_MODEL": (
|
|
"meta-llama/Llama-3.2-3B-Instruct",
|
|
"Inference model loaded into the vLLM engine",
|
|
),
|
|
"TENSOR_PARALLEL_SIZE": (
|
|
"1",
|
|
"Number of tensor parallel replicas (number of GPUs to use).",
|
|
),
|
|
"MAX_TOKENS": (
|
|
"4096",
|
|
"Maximum number of tokens to generate.",
|
|
),
|
|
"ENFORCE_EAGER": (
|
|
"False",
|
|
"Whether to use eager mode for inference (otherwise cuda graphs are used).",
|
|
),
|
|
"GPU_MEMORY_UTILIZATION": (
|
|
"0.7",
|
|
"GPU memory utilization for the vLLM engine.",
|
|
),
|
|
},
|
|
)
|