chore: move all Llama Stack types from llama-models to llama-stack (#1098)

llama-models should have extremely minimal cruft. Its sole purpose
should be didactic -- show the simplest implementation of the llama
models and document the prompt formats, etc.

This PR is the complement to
https://github.com/meta-llama/llama-models/pull/279

## Test Plan

Ensure all `llama` CLI `model` sub-commands work:

```bash
llama model list
llama model download --model-id ...
llama model prompt-format -m ...
```

Ran tests:
```bash
cd tests/client-sdk
LLAMA_STACK_CONFIG=fireworks pytest -s -v inference/
LLAMA_STACK_CONFIG=fireworks pytest -s -v vector_io/
LLAMA_STACK_CONFIG=fireworks pytest -s -v agents/
```

Create a fresh venv `uv venv && source .venv/bin/activate` and run
`llama stack build --template fireworks --image-type venv` followed by
`llama stack run together --image-type venv` <-- the server runs

Also checked that the OpenAPI generator can run and there is no change
in the generated files as a result.

```bash
cd docs/openapi_generator
sh run_openapi_generator.sh
```

llama_stack/providers/remote/inference/bedrock/bedrock.py

@@ -8,7 +8,6 @@ import json
 from typing import AsyncGenerator, AsyncIterator, Dict, List, Optional, Union
 
 from botocore.client import BaseClient
-from llama_models.datatypes import CoreModelId
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.tokenizer import Tokenizer
 
@@ -28,6 +27,7 @@ from llama_stack.apis.inference import (
     ToolDefinition,
     ToolPromptFormat,
 )
+from llama_stack.models.llama.datatypes import CoreModelId
 from llama_stack.providers.remote.inference.bedrock.config import BedrockConfig
 from llama_stack.providers.utils.bedrock.client import create_bedrock_client
 from llama_stack.providers.utils.inference.model_registry import (

llama_stack/providers/remote/inference/cerebras/cerebras.py

@@ -7,9 +7,7 @@
 from typing import AsyncGenerator, List, Optional, Union
 
 from cerebras.cloud.sdk import AsyncCerebras
-from llama_models.datatypes import CoreModelId
 from llama_models.llama3.api.chat_format import ChatFormat
-from llama_models.llama3.api.datatypes import TopKSamplingStrategy
 from llama_models.llama3.api.tokenizer import Tokenizer
 
 from llama_stack.apis.common.content_types import InterleavedContent
@@ -28,6 +26,7 @@ from llama_stack.apis.inference import (
     ToolDefinition,
     ToolPromptFormat,
 )
+from llama_stack.models.llama.datatypes import CoreModelId, TopKSamplingStrategy
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
     build_model_alias,

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

@@ -7,9 +7,10 @@
 import os
 from typing import Any, Dict, Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field, SecretStr
 
+from llama_stack.schema_utils import json_schema_type
+
 DEFAULT_BASE_URL = "https://api.cerebras.ai"
 
 

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

@@ -5,9 +5,10 @@
 # the root directory of this source tree.
 
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class DatabricksImplConfig(BaseModel):

llama_stack/providers/remote/inference/databricks/databricks.py

@@ -6,7 +6,6 @@
 
 from typing import AsyncGenerator, List, Optional
 
-from llama_models.datatypes import CoreModelId
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.tokenizer import Tokenizer
 from openai import OpenAI
@@ -25,6 +24,7 @@ from llama_stack.apis.inference import (
     ToolDefinition,
     ToolPromptFormat,
 )
+from llama_stack.models.llama.datatypes import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
     build_model_alias,

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

@@ -6,9 +6,10 @@
 
 from typing import Any, Dict, Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field, SecretStr
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class FireworksImplConfig(BaseModel):

llama_stack/providers/remote/inference/fireworks/fireworks.py

@@ -7,7 +7,6 @@
 from typing import AsyncGenerator, List, Optional, Union
 
 from fireworks.client import Fireworks
-from llama_models.datatypes import CoreModelId
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.tokenizer import Tokenizer
 
@@ -30,6 +29,7 @@ from llama_stack.apis.inference import (
     ToolPromptFormat,
 )
 from llama_stack.distribution.request_headers import NeedsRequestProviderData
+from llama_stack.models.llama.datatypes import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
     build_model_alias,

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

@@ -6,9 +6,10 @@
 
 from typing import Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class GroqConfig(BaseModel):

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

@@ -9,9 +9,6 @@ from typing import AsyncIterator, List, Optional, Union
 
 import groq
 from groq import Groq
-from llama_models.datatypes import SamplingParams
-from llama_models.llama3.api.datatypes import ToolDefinition, ToolPromptFormat
-from llama_models.sku_list import CoreModelId
 
 from llama_stack.apis.inference import (
     ChatCompletionRequest,
@@ -29,6 +26,8 @@ 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.providers.remote.inference.groq.config import GroqConfig
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,

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

@@ -24,7 +24,6 @@ from groq.types.chat.chat_completion_user_message_param import (
 )
 from groq.types.chat.completion_create_params import CompletionCreateParams
 from groq.types.shared.function_definition import FunctionDefinition
-from llama_models.llama3.api.datatypes import ToolParamDefinition
 
 from llama_stack.apis.common.content_types import (
     TextDelta,
@@ -44,6 +43,7 @@ from llama_stack.apis.inference import (
     ToolDefinition,
     ToolPromptFormat,
 )
+from llama_stack.models.llama.datatypes import ToolParamDefinition
 from llama_stack.providers.utils.inference.openai_compat import (
     UnparseableToolCall,
     convert_tool_call,

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

@@ -7,9 +7,10 @@
 import os
 from typing import Any, Dict, Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field, SecretStr
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class NVIDIAConfig(BaseModel):

llama_stack/providers/remote/inference/nvidia/nvidia.py

@@ -7,9 +7,6 @@
 import warnings
 from typing import AsyncIterator, List, Optional, Union
 
-from llama_models.datatypes import SamplingParams
-from llama_models.llama3.api.datatypes import ToolDefinition, ToolPromptFormat
-from llama_models.sku_list import CoreModelId
 from openai import APIConnectionError, AsyncOpenAI
 
 from llama_stack.apis.inference import (
@@ -28,6 +25,7 @@ from llama_stack.apis.inference import (
     ToolChoice,
     ToolConfig,
 )
+from llama_stack.models.llama.datatypes import CoreModelId, SamplingParams, ToolDefinition, ToolPromptFormat
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
     build_model_alias,

llama_stack/providers/remote/inference/nvidia/openai_utils.py

@@ -8,17 +8,6 @@ import json
 import warnings
 from typing import Any, AsyncGenerator, Dict, Generator, Iterable, List, Optional, Union
 
-from llama_models.datatypes import (
-    GreedySamplingStrategy,
-    TopKSamplingStrategy,
-    TopPSamplingStrategy,
-)
-from llama_models.llama3.api.datatypes import (
-    BuiltinTool,
-    StopReason,
-    ToolCall,
-    ToolDefinition,
-)
 from openai import AsyncStream
 from openai.types.chat import (
     ChatCompletionAssistantMessageParam as OpenAIChatCompletionAssistantMessage,
@@ -87,6 +76,15 @@ from llama_stack.apis.inference import (
     ToolResponseMessage,
     UserMessage,
 )
+from llama_stack.models.llama.datatypes import (
+    BuiltinTool,
+    GreedySamplingStrategy,
+    StopReason,
+    ToolCall,
+    ToolDefinition,
+    TopKSamplingStrategy,
+    TopPSamplingStrategy,
+)
 from llama_stack.providers.utils.inference.prompt_adapter import (
     convert_image_content_to_url,
 )

llama_stack/providers/remote/inference/ollama/ollama.py

@@ -8,7 +8,6 @@ import logging
 from typing import AsyncGenerator, List, Optional, Union
 
 import httpx
-from llama_models.datatypes import CoreModelId
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.tokenizer import Tokenizer
 from ollama import AsyncClient
@@ -34,6 +33,7 @@ from llama_stack.apis.inference import (
     ToolPromptFormat,
 )
 from llama_stack.apis.models import Model, ModelType
+from llama_stack.models.llama.datatypes import CoreModelId
 from llama_stack.providers.datatypes import ModelsProtocolPrivate
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,

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

@@ -6,9 +6,10 @@
 
 from typing import Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class RunpodImplConfig(BaseModel):

llama_stack/providers/remote/inference/runpod/runpod.py

@@ -6,11 +6,11 @@
 from typing import AsyncGenerator
 
 from llama_models.llama3.api.chat_format import ChatFormat
-from llama_models.llama3.api.datatypes import Message
 from llama_models.llama3.api.tokenizer import Tokenizer
 from openai import OpenAI
 
 from llama_stack.apis.inference import *  # noqa: F403
+from llama_stack.models.llama.datatypes import Message
 
 # from llama_stack.providers.datatypes import ModelsProtocolPrivate
 from llama_stack.providers.utils.inference.model_registry import ModelRegistryHelper

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

@@ -6,9 +6,10 @@
 
 from typing import Any, Dict, Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class SambaNovaImplConfig(BaseModel):

llama_stack/providers/remote/inference/sambanova/sambanova.py

@@ -7,12 +7,6 @@
 import json
 from typing import AsyncGenerator
 
-from llama_models.datatypes import (
-    CoreModelId,
-    GreedySamplingStrategy,
-    TopKSamplingStrategy,
-    TopPSamplingStrategy,
-)
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.tokenizer import Tokenizer
 from openai import OpenAI
@@ -23,6 +17,12 @@ from llama_stack.apis.common.content_types import (
     TextContentItem,
 )
 from llama_stack.apis.inference import *  # noqa: F403
+from llama_stack.models.llama.datatypes import (
+    CoreModelId,
+    GreedySamplingStrategy,
+    TopKSamplingStrategy,
+    TopPSamplingStrategy,
+)
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
     build_model_alias,

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

@@ -6,9 +6,10 @@
 
 from typing import Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field, SecretStr
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class TGIImplConfig(BaseModel):

llama_stack/providers/remote/inference/tgi/tgi.py

@@ -11,7 +11,6 @@ from typing import AsyncGenerator, List, Optional
 from huggingface_hub import AsyncInferenceClient, HfApi
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.tokenizer import Tokenizer
-from llama_models.sku_list import all_registered_models
 
 from llama_stack.apis.common.content_types import InterleavedContent
 from llama_stack.apis.inference import (
@@ -31,6 +30,7 @@ from llama_stack.apis.inference import (
     ToolPromptFormat,
 )
 from llama_stack.apis.models import Model
+from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.datatypes import ModelsProtocolPrivate
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,

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

@@ -6,9 +6,10 @@
 
 from typing import Any, Dict, Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field, SecretStr
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class TogetherImplConfig(BaseModel):

llama_stack/providers/remote/inference/together/together.py

@@ -6,7 +6,6 @@
 
 from typing import AsyncGenerator, List, Optional, Union
 
-from llama_models.datatypes import CoreModelId
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.tokenizer import Tokenizer
 from together import Together
@@ -29,6 +28,7 @@ from llama_stack.apis.inference import (
     ToolPromptFormat,
 )
 from llama_stack.distribution.request_headers import NeedsRequestProviderData
+from llama_stack.models.llama.datatypes import CoreModelId
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,
     build_model_alias,

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

@@ -6,9 +6,10 @@
 
 from typing import Optional
 
-from llama_models.schema_utils import json_schema_type
 from pydantic import BaseModel, Field
 
+from llama_stack.schema_utils import json_schema_type
+
 
 @json_schema_type
 class VLLMInferenceAdapterConfig(BaseModel):

llama_stack/providers/remote/inference/vllm/vllm.py

@@ -7,10 +7,9 @@ import json
 import logging
 from typing import AsyncGenerator, List, Optional, Union
 
-from llama_models.llama3.api import StopReason, ToolCall
+from llama_models.datatypes import StopReason, ToolCall
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.tokenizer import Tokenizer
-from llama_models.sku_list import all_registered_models
 from openai import OpenAI
 
 from llama_stack.apis.common.content_types import InterleavedContent, TextDelta, ToolCallDelta, ToolCallParseStatus
@@ -37,6 +36,7 @@ from llama_stack.apis.inference import (
     ToolPromptFormat,
 )
 from llama_stack.apis.models import Model, ModelType
+from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.datatypes import ModelsProtocolPrivate
 from llama_stack.providers.utils.inference.model_registry import (
     ModelRegistryHelper,