forked from phoenix-oss/llama-stack-mirror
252a487085
# What does this PR do? Adds nvidia as a safety provider by interfacing with the nemo guardrails microservice. This enables checking user’s input or the LLM’s output against input and output guardrails by using the `/v1/guardrails/checks` endpoint of the[ guardrails API.](https://developer.nvidia.com/docs/nemo-microservices/guardrails/source/guides/checks-guide.html) ## Test Plan Deploy nemo guardrails service following the documentation: https://developer.nvidia.com/docs/nemo-microservices/guardrails/source/getting-started/deploy-docker.html ### Standalone: ```bash (venv) local-cdgamarose@a1u1g-rome-0153:~/llama-stack$ pytest -v -s llama_stack/providers/tests/safety/test_safety.py --providers inference=nvidia,safety=nvidia --safety-shield meta/llama-3.1-8b-instruct =================================================================================== test session starts =================================================================================== platform linux -- Python 3.10.12, pytest-8.3.4, pluggy-1.5.0 -- /localhome/local-cdgamarose/llama-stack/venv/bin/python3 cachedir: .pytest_cache metadata: {'Python': '3.10.12', 'Platform': 'Linux-5.15.0-122-generic-x86_64-with-glibc2.35', 'Packages': {'pytest': '8.3.4', 'pluggy': '1.5.0'}, 'Plugins': {'metadata': '3.1.1', 'asyncio': '0.25.3', 'anyio': '4.8.0', 'html': '4.1.1'}} rootdir: /localhome/local-cdgamarose/llama-stack configfile: pyproject.toml plugins: metadata-3.1.1, asyncio-0.25.3, anyio-4.8.0, html-4.1.1 asyncio: mode=strict, asyncio_default_fixture_loop_scope=None collected 2 items llama_stack/providers/tests/safety/test_safety.py::TestSafety::test_shield_list[--inference=nvidia:safety=nvidia] Initializing NVIDIASafetyAdapter(http://0.0.0.0:7331)... PASSED llama_stack/providers/tests/safety/test_safety.py::TestSafety::test_run_shield[--inference=nvidia:safety=nvidia] PASSED ============================================================================== 2 passed, 2 warnings in 4.78s ============================================================================== ``` ### Distribution: ``` llama stack run llama_stack/templates/nvidia/run-with-safety.yaml curl -v -X 'POST' "http://localhost:8321/v1/safety/run-shield" -H 'accept: application/json' -H 'Content-Type: application/json' -d '{"shield_id": "meta/llama-3.1-8b-instruct", "messages":[{"role": "user", "content": "you are stupid"}]}' {"violation":{"violation_level":"error","user_message":"Sorry I cannot do this.","metadata":{"self check input":{"status":"blocked"}}}} ``` [//]: # (## Documentation) --------- Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
105 lines
3.6 KiB
Python
105 lines
3.6 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"],
|
|
"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",
|
|
),
|
|
"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",
|
|
),
|
|
},
|
|
)
|