feat: Add Groq distribution template (#1173)

# What does this PR do?

Create a distribution template using Groq as inference provider.
Link to issue: https://github.com/meta-llama/llama-stack/issues/958


## Test Plan
Run `python llama_stack/scripts/distro_codegen.py` to generate run.yaml
and build.yaml
Test the newly created template by running
`llama stack build --template <template-name>`
`llama stack run <template-name>`

distributions/dependencies.json

@@ -171,6 +171,39 @@
     "sentence-transformers --no-deps",
     "torch torchvision --index-url https://download.pytorch.org/whl/cpu"
   ],
+  "groq": [
+    "aiosqlite",
+    "autoevals",
+    "blobfile",
+    "chardet",
+    "datasets",
+    "faiss-cpu",
+    "fastapi",
+    "fire",
+    "groq",
+    "httpx",
+    "matplotlib",
+    "nltk",
+    "numpy",
+    "openai",
+    "opentelemetry-exporter-otlp-proto-http",
+    "opentelemetry-sdk",
+    "pandas",
+    "pillow",
+    "psycopg2-binary",
+    "pymongo",
+    "pypdf",
+    "redis",
+    "requests",
+    "scikit-learn",
+    "scipy",
+    "sentencepiece",
+    "tqdm",
+    "transformers",
+    "uvicorn",
+    "sentence-transformers --no-deps",
+    "torch torchvision --index-url https://download.pytorch.org/whl/cpu"
+  ],
   "hf-endpoint": [
     "aiohttp",
     "aiosqlite",

docs/source/distributions/self_hosted_distro/groq.md

@@ -0,0 +1,77 @@
+---
+orphan: true
+---
+<!-- This file was auto-generated by distro_codegen.py, please edit source -->
+# Groq Distribution
+
+```{toctree}
+:maxdepth: 2
+:hidden:
+
+self
+```
+
+The `llamastack/distribution-groq` distribution consists of the following provider configurations.
+
+| API | Provider(s) |
+|-----|-------------|
+| agents | `inline::meta-reference` |
+| datasetio | `remote::huggingface`, `inline::localfs` |
+| eval | `inline::meta-reference` |
+| inference | `remote::groq` |
+| safety | `inline::llama-guard` |
+| scoring | `inline::basic`, `inline::llm-as-judge`, `inline::braintrust` |
+| telemetry | `inline::meta-reference` |
+| tool_runtime | `remote::brave-search`, `remote::tavily-search`, `inline::code-interpreter`, `inline::rag-runtime` |
+| vector_io | `inline::faiss` |
+
+
+### Environment Variables
+
+The following environment variables can be configured:
+
+- `LLAMASTACK_PORT`: Port for the Llama Stack distribution server (default: `5001`)
+- `GROQ_API_KEY`: Groq API Key (default: ``)
+
+### Models
+
+The following models are available by default:
+
+- `meta-llama/Llama-3.1-8B-Instruct (llama3-8b-8192)`
+- `meta-llama/Llama-3.1-8B-Instruct (llama-3.1-8b-instant)`
+- `meta-llama/Llama-3-70B-Instruct (llama3-70b-8192)`
+- `meta-llama/Llama-3.3-70B-Instruct (llama-3.3-70b-versatile)`
+- `meta-llama/Llama-3.2-3B-Instruct (llama-3.2-3b-preview)`
+
+
+### Prerequisite: API Keys
+
+Make sure you have access to a Groq API Key. You can get one by visiting [Groq](https://api.groq.com/).
+
+
+## Running Llama Stack with Groq
+
+You can do this via Conda (build code) or Docker which has a pre-built image.
+
+### Via Docker
+
+This method allows you to get started quickly without having to build the distribution code.
+
+```bash
+LLAMA_STACK_PORT=5001
+docker run \
+  -it \
+  -p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
+  llamastack/distribution-groq \
+  --port $LLAMA_STACK_PORT \
+  --env GROQ_API_KEY=$GROQ_API_KEY
+```
+
+### Via Conda
+
+```bash
+llama stack build --template groq --image-type conda
+llama stack run ./run.yaml \
+  --port $LLAMA_STACK_PORT \
+  --env GROQ_API_KEY=$GROQ_API_KEY
+```

llama_stack/providers/remote/inference/groq/config.py

@@ -4,7 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from typing import Optional
+from typing import Any, Dict, Optional
 
 from pydantic import BaseModel, Field
 
@@ -18,3 +18,15 @@ class GroqConfig(BaseModel):
         default=None,
         description="The Groq API key",
     )
+
+    url: str = Field(
+        default="https://api.groq.com",
+        description="The URL for the Groq AI server",
+    )
+
+    @classmethod
+    def sample_run_config(cls, **kwargs) -> Dict[str, Any]:
+        return {
+            "url": "https://api.groq.com",
+            "api_key": "${env.GROQ_API_KEY}",
+        }

llama_stack/providers/remote/inference/groq/groq.py

@@ -29,17 +29,10 @@ from llama_stack.apis.inference import (
     ToolConfig,
 )
 from llama_stack.distribution.request_headers import NeedsRequestProviderData
-from llama_stack.models.llama.datatypes import (
-    SamplingParams,
-    ToolDefinition,
-    ToolPromptFormat,
-)
-from llama_stack.models.llama.sku_list import CoreModelId
+from llama_stack.models.llama.datatypes import SamplingParams, ToolDefinition, ToolPromptFormat
 from llama_stack.providers.remote.inference.groq.config import GroqConfig
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
-    build_hf_repo_model_entry,
-    build_model_entry,
 )
 
 from .groq_utils import (
@@ -47,33 +40,7 @@ from .groq_utils import (
     convert_chat_completion_response,
     convert_chat_completion_response_stream,
 )
-
-_MODEL_ENTRIES = [
-    build_hf_repo_model_entry(
-        "llama3-8b-8192",
-        CoreModelId.llama3_1_8b_instruct.value,
-    ),
-    build_model_entry(
-        "llama-3.1-8b-instant",
-        CoreModelId.llama3_1_8b_instruct.value,
-    ),
-    build_hf_repo_model_entry(
-        "llama3-70b-8192",
-        CoreModelId.llama3_70b_instruct.value,
-    ),
-    build_hf_repo_model_entry(
-        "llama-3.3-70b-versatile",
-        CoreModelId.llama3_3_70b_instruct.value,
-    ),
-    # Groq only contains a preview version for llama-3.2-3b
-    # Preview models aren't recommended for production use, but we include this one
-    # to pass the test fixture
-    # TODO(aidand): Replace this with a stable model once Groq supports it
-    build_hf_repo_model_entry(
-        "llama-3.2-3b-preview",
-        CoreModelId.llama3_2_3b_instruct.value,
-    ),
-]
+from .models import _MODEL_ENTRIES
 
 
 class GroqInferenceAdapter(Inference, ModelRegistryHelper, NeedsRequestProviderData):

llama_stack/providers/remote/inference/groq/models.py

@@ -0,0 +1,35 @@
+# 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.models.llama.sku_list import CoreModelId
+from llama_stack.providers.utils.inference.model_registry import build_model_entry
+
+_MODEL_ENTRIES = [
+    build_model_entry(
+        "llama3-8b-8192",
+        CoreModelId.llama3_1_8b_instruct.value,
+    ),
+    build_model_entry(
+        "llama-3.1-8b-instant",
+        CoreModelId.llama3_1_8b_instruct.value,
+    ),
+    build_model_entry(
+        "llama3-70b-8192",
+        CoreModelId.llama3_70b_instruct.value,
+    ),
+    build_model_entry(
+        "llama-3.3-70b-versatile",
+        CoreModelId.llama3_3_70b_instruct.value,
+    ),
+    # Groq only contains a preview version for llama-3.2-3b
+    # Preview models aren't recommended for production use, but we include this one
+    # to pass the test fixture
+    # TODO(aidand): Replace this with a stable model once Groq supports it
+    build_model_entry(
+        "llama-3.2-3b-preview",
+        CoreModelId.llama3_2_3b_instruct.value,
+    ),
+]

llama_stack/templates/groq/__init__.py

@@ -0,0 +1,7 @@
+# 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 .groq import get_distribution_template  # noqa: F401

llama_stack/templates/groq/build.yaml

@@ -0,0 +1,29 @@
+version: '2'
+distribution_spec:
+  description: Use Groq for running LLM inference
+  providers:
+    inference:
+    - remote::groq
+    vector_io:
+    - inline::faiss
+    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
+image_type: conda

llama_stack/templates/groq/doc_template.md

@@ -0,0 +1,68 @@
+---
+orphan: true
+---
+# Groq Distribution
+
+```{toctree}
+:maxdepth: 2
+:hidden:
+
+self
+```
+
+The `llamastack/distribution-{{ name }}` distribution consists of the following provider configurations.
+
+{{ providers_table }}
+
+{% if run_config_env_vars %}
+### Environment Variables
+
+The following environment variables can be configured:
+
+{% for var, (default_value, description) in run_config_env_vars.items() %}
+- `{{ var }}`: {{ description }} (default: `{{ default_value }}`)
+{% endfor %}
+{% endif %}
+
+{% if default_models %}
+### Models
+
+The following models are available by default:
+
+{% for model in default_models %}
+- `{{ model.model_id }} ({{ model.provider_model_id }})`
+{% endfor %}
+{% endif %}
+
+
+### Prerequisite: API Keys
+
+Make sure you have access to a Groq API Key. You can get one by visiting [Groq](https://api.groq.com/).
+
+
+## Running Llama Stack with Groq
+
+You can do this via Conda (build code) or Docker which has a pre-built image.
+
+### Via Docker
+
+This method allows you to get started quickly without having to build the distribution code.
+
+```bash
+LLAMA_STACK_PORT=5001
+docker run \
+  -it \
+  -p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
+  llamastack/distribution-{{ name }} \
+  --port $LLAMA_STACK_PORT \
+  --env GROQ_API_KEY=$GROQ_API_KEY
+```
+
+### Via Conda
+
+```bash
+llama stack build --template groq --image-type conda
+llama stack run ./run.yaml \
+  --port $LLAMA_STACK_PORT \
+  --env GROQ_API_KEY=$GROQ_API_KEY
+```

llama_stack/templates/groq/groq.py

@@ -0,0 +1,121 @@
+# 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.apis.models.models import ModelType
+from llama_stack.distribution.datatypes import (
+    ModelInput,
+    Provider,
+    ToolGroupInput,
+)
+from llama_stack.models.llama.sku_list import all_registered_models
+from llama_stack.providers.inline.inference.sentence_transformers import (
+    SentenceTransformersInferenceConfig,
+)
+from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
+from llama_stack.providers.remote.inference.groq import GroqConfig
+from llama_stack.providers.remote.inference.groq.models import _MODEL_ENTRIES
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+
+
+def get_distribution_template() -> DistributionTemplate:
+    providers = {
+        "inference": ["remote::groq"],
+        "vector_io": ["inline::faiss"],
+        "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",
+        ],
+    }
+    name = "groq"
+
+    inference_provider = Provider(
+        provider_id=name,
+        provider_type=f"remote::{name}",
+        config=GroqConfig.sample_run_config(),
+    )
+
+    embedding_provider = Provider(
+        provider_id="sentence-transformers",
+        provider_type="inline::sentence-transformers",
+        config=SentenceTransformersInferenceConfig.sample_run_config(),
+    )
+    vector_io_provider = Provider(
+        provider_id="faiss",
+        provider_type="inline::faiss",
+        config=FaissVectorIOConfig.sample_run_config(f"distributions/{name}"),
+    )
+    embedding_model = ModelInput(
+        model_id="all-MiniLM-L6-v2",
+        provider_id="sentence-transformers",
+        model_type=ModelType.embedding,
+        metadata={
+            "embedding_dimension": 384,
+        },
+    )
+
+    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+    default_models = [
+        ModelInput(
+            model_id=core_model_to_hf_repo[m.llama_model],
+            provider_model_id=m.provider_model_id,
+            provider_id=name,
+        )
+        for m in _MODEL_ENTRIES
+    ]
+
+    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 Groq for running LLM inference",
+        docker_image=None,
+        template_path=Path(__file__).parent / "doc_template.md",
+        providers=providers,
+        default_models=default_models,
+        run_configs={
+            "run.yaml": RunConfigSettings(
+                provider_overrides={
+                    "inference": [inference_provider, embedding_provider],
+                },
+                default_models=default_models + [embedding_model],
+                default_tool_groups=default_tool_groups,
+            ),
+        },
+        run_config_env_vars={
+            "LLAMASTACK_PORT": (
+                "5001",
+                "Port for the Llama Stack distribution server",
+            ),
+            "GROQ_API_KEY": (
+                "",
+                "Groq API Key",
+            ),
+        },
+    )

llama_stack/templates/groq/run.yaml

@@ -0,0 +1,136 @@
+version: '2'
+image_name: groq
+apis:
+- agents
+- datasetio
+- eval
+- inference
+- safety
+- scoring
+- telemetry
+- tool_runtime
+- vector_io
+providers:
+  inference:
+  - provider_id: groq
+    provider_type: remote::groq
+    config:
+      url: https://api.groq.com
+      api_key: ${env.GROQ_API_KEY}
+  - provider_id: sentence-transformers
+    provider_type: inline::sentence-transformers
+    config: {}
+  vector_io:
+  - provider_id: faiss
+    provider_type: inline::faiss
+    config:
+      kvstore:
+        type: sqlite
+        namespace: null
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/faiss_store.db
+  safety:
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
+    config: {}
+  agents:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config:
+      persistence_store:
+        type: sqlite
+        namespace: null
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/agents_store.db
+  telemetry:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config:
+      service_name: ${env.OTEL_SERVICE_NAME:llama-stack}
+      sinks: ${env.TELEMETRY_SINKS:console,sqlite}
+      sqlite_db_path: ${env.SQLITE_DB_PATH:~/.llama/distributions/groq/trace_store.db}
+  eval:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config: {}
+  datasetio:
+  - provider_id: huggingface
+    provider_type: remote::huggingface
+    config: {}
+  - provider_id: localfs
+    provider_type: inline::localfs
+    config: {}
+  scoring:
+  - provider_id: basic
+    provider_type: inline::basic
+    config: {}
+  - provider_id: llm-as-judge
+    provider_type: inline::llm-as-judge
+    config: {}
+  - provider_id: braintrust
+    provider_type: inline::braintrust
+    config:
+      openai_api_key: ${env.OPENAI_API_KEY:}
+  tool_runtime:
+  - provider_id: brave-search
+    provider_type: remote::brave-search
+    config:
+      api_key: ${env.BRAVE_SEARCH_API_KEY:}
+      max_results: 3
+  - provider_id: tavily-search
+    provider_type: remote::tavily-search
+    config:
+      api_key: ${env.TAVILY_SEARCH_API_KEY:}
+      max_results: 3
+  - provider_id: code-interpreter
+    provider_type: inline::code-interpreter
+    config: {}
+  - provider_id: rag-runtime
+    provider_type: inline::rag-runtime
+    config: {}
+metadata_store:
+  type: sqlite
+  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/registry.db
+models:
+- metadata: {}
+  model_id: meta-llama/Llama-3.1-8B-Instruct
+  provider_id: groq
+  provider_model_id: llama3-8b-8192
+  model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.1-8B-Instruct
+  provider_id: groq
+  provider_model_id: llama-3.1-8b-instant
+  model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3-70B-Instruct
+  provider_id: groq
+  provider_model_id: llama3-70b-8192
+  model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.3-70B-Instruct
+  provider_id: groq
+  provider_model_id: llama-3.3-70b-versatile
+  model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-3B-Instruct
+  provider_id: groq
+  provider_model_id: llama-3.2-3b-preview
+  model_type: llm
+- metadata:
+    embedding_dimension: 384
+  model_id: all-MiniLM-L6-v2
+  provider_id: sentence-transformers
+  model_type: embedding
+shields: []
+vector_dbs: []
+datasets: []
+scoring_fns: []
+benchmarks: []
+tool_groups:
+- toolgroup_id: builtin::websearch
+  provider_id: tavily-search
+- toolgroup_id: builtin::rag
+  provider_id: rag-runtime
+- toolgroup_id: builtin::code_interpreter
+  provider_id: code-interpreter
+server:
+  port: 8321