mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-06-27 18:50:41 +00:00
1a73f8305b
# What does this PR do? This PR adds support for NVIDIA's NeMo Customizer API to the Llama Stack post-training module. The integration enables users to fine-tune models using NVIDIA's cloud-based customization service through a consistent Llama Stack interface. [//]: # (If resolving an issue, uncomment and update the line below) [//]: # (Closes #[issue-number]) ## Test Plan [Describe the tests you ran to verify your changes with result summaries. *Provide clear instructions so the plan can be easily re-executed.*] Yet to be done Things pending under this PR: - [x] Integration of fine-tuned model(new checkpoint) for inference with nvidia llm distribution - [x] distribution integration of API - [x] Add test cases for customizer(In Progress) - [x] Documentation ``` LLAMA_STACK_BASE_URL=http://localhost:5002 pytest -v tests/client-sdk/post_training/test_supervised_fine_tuning.py ============================================================================================================================================================================ test session starts ============================================================================================================================================================================= platform linux -- Python 3.10.0, pytest-8.3.4, pluggy-1.5.0 -- /home/ubuntu/llama-stack/.venv/bin/python cachedir: .pytest_cache metadata: {'Python': '3.10.0', 'Platform': 'Linux-6.8.0-1021-gcp-x86_64-with-glibc2.35', 'Packages': {'pytest': '8.3.4', 'pluggy': '1.5.0'}, 'Plugins': {'nbval': '0.11.0', 'metadata': '3.1.1', 'anyio': '4.8.0', 'html': '4.1.1', 'asyncio': '0.25.3'}} rootdir: /home/ubuntu/llama-stack configfile: pyproject.toml plugins: nbval-0.11.0, metadata-3.1.1, anyio-4.8.0, html-4.1.1, asyncio-0.25.3 asyncio: mode=strict, asyncio_default_fixture_loop_scope=None collected 2 items tests/client-sdk/post_training/test_supervised_fine_tuning.py::test_post_training_provider_registration[txt=8B] PASSED [ 50%] tests/client-sdk/post_training/test_supervised_fine_tuning.py::test_list_training_jobs[txt=8B] PASSED [100%] ======================================================================================================================================================================== 2 passed, 1 warning in 0.10s ======================================================================================================================================================================== ``` cc: @mattf @dglogo @sumitb --------- Co-authored-by: Ubuntu <ubuntu@llama-stack-customizer-dev-inst-2tx95fyisatvlic4we8hidx5tfj.us-central1-a.c.brevdevprod.internal>
131 lines
4.4 KiB
Python
131 lines
4.4 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 pathlib import Path
|
|
|
|
from llama_stack.distribution.datatypes import ModelInput, Provider, ShieldInput, ToolGroupInput
|
|
from llama_stack.providers.remote.inference.nvidia import NVIDIAConfig
|
|
from llama_stack.providers.remote.inference.nvidia.models import MODEL_ENTRIES
|
|
from llama_stack.providers.remote.safety.nvidia import NVIDIASafetyConfig
|
|
from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
|
|
|
|
|
|
def get_distribution_template() -> DistributionTemplate:
|
|
providers = {
|
|
"inference": ["remote::nvidia"],
|
|
"vector_io": ["inline::faiss"],
|
|
"safety": ["remote::nvidia"],
|
|
"agents": ["inline::meta-reference"],
|
|
"telemetry": ["inline::meta-reference"],
|
|
"eval": ["inline::meta-reference"],
|
|
"post_training": ["remote::nvidia"],
|
|
"datasetio": ["inline::localfs"],
|
|
"scoring": ["inline::basic"],
|
|
"tool_runtime": ["inline::rag-runtime"],
|
|
}
|
|
|
|
inference_provider = Provider(
|
|
provider_id="nvidia",
|
|
provider_type="remote::nvidia",
|
|
config=NVIDIAConfig.sample_run_config(),
|
|
)
|
|
safety_provider = Provider(
|
|
provider_id="nvidia",
|
|
provider_type="remote::nvidia",
|
|
config=NVIDIASafetyConfig.sample_run_config(),
|
|
)
|
|
inference_model = ModelInput(
|
|
model_id="${env.INFERENCE_MODEL}",
|
|
provider_id="nvidia",
|
|
)
|
|
safety_model = ModelInput(
|
|
model_id="${env.SAFETY_MODEL}",
|
|
provider_id="nvidia",
|
|
)
|
|
|
|
available_models = {
|
|
"nvidia": MODEL_ENTRIES,
|
|
}
|
|
default_tool_groups = [
|
|
ToolGroupInput(
|
|
toolgroup_id="builtin::rag",
|
|
provider_id="rag-runtime",
|
|
),
|
|
]
|
|
|
|
default_models = get_model_registry(available_models)
|
|
return DistributionTemplate(
|
|
name="nvidia",
|
|
distro_type="remote_hosted",
|
|
description="Use NVIDIA NIM for running LLM inference and safety",
|
|
container_image=None,
|
|
template_path=Path(__file__).parent / "doc_template.md",
|
|
providers=providers,
|
|
available_models_by_provider=available_models,
|
|
run_configs={
|
|
"run.yaml": RunConfigSettings(
|
|
provider_overrides={
|
|
"inference": [inference_provider],
|
|
},
|
|
default_models=default_models,
|
|
default_tool_groups=default_tool_groups,
|
|
),
|
|
"run-with-safety.yaml": RunConfigSettings(
|
|
provider_overrides={
|
|
"inference": [
|
|
inference_provider,
|
|
safety_provider,
|
|
]
|
|
},
|
|
default_models=[inference_model, safety_model],
|
|
default_shields=[ShieldInput(shield_id="${env.SAFETY_MODEL}", provider_id="nvidia")],
|
|
default_tool_groups=default_tool_groups,
|
|
),
|
|
},
|
|
run_config_env_vars={
|
|
"NVIDIA_API_KEY": (
|
|
"",
|
|
"NVIDIA API Key",
|
|
),
|
|
## Nemo Customizer related variables
|
|
"NVIDIA_USER_ID": (
|
|
"llama-stack-user",
|
|
"NVIDIA User ID",
|
|
),
|
|
"NVIDIA_DATASET_NAMESPACE": (
|
|
"default",
|
|
"NVIDIA Dataset Namespace",
|
|
),
|
|
"NVIDIA_ACCESS_POLICIES": (
|
|
"{}",
|
|
"NVIDIA Access Policies",
|
|
),
|
|
"NVIDIA_PROJECT_ID": (
|
|
"test-project",
|
|
"NVIDIA Project ID",
|
|
),
|
|
"NVIDIA_CUSTOMIZER_URL": (
|
|
"https://customizer.api.nvidia.com",
|
|
"NVIDIA Customizer URL",
|
|
),
|
|
"NVIDIA_OUTPUT_MODEL_DIR": (
|
|
"test-example-model@v1",
|
|
"NVIDIA Output Model Directory",
|
|
),
|
|
"GUARDRAILS_SERVICE_URL": (
|
|
"http://0.0.0.0:7331",
|
|
"URL for the NeMo Guardrails Service",
|
|
),
|
|
"INFERENCE_MODEL": (
|
|
"Llama3.1-8B-Instruct",
|
|
"Inference model",
|
|
),
|
|
"SAFETY_MODEL": (
|
|
"meta/llama-3.1-8b-instruct",
|
|
"Name of the model to use for safety",
|
|
),
|
|
},
|
|
)
|