121 lines
3.1 KiB
Python
121 lines
3.1 KiB
Python
"""Models endpoint router."""
|
|
|
|
import time
|
|
from fastapi import APIRouter
|
|
from app.models.openai_models import ModelsResponse, ModelInfo
|
|
from app.config import settings
|
|
import logging
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
router = APIRouter()
|
|
|
|
|
|
# Predefined list of available models
|
|
# This can be extended or made dynamic based on watsonx.ai API
|
|
AVAILABLE_MODELS = [
|
|
# Granite Models
|
|
"ibm/granite-3-1-8b-base",
|
|
"ibm/granite-3-2-8b-instruct",
|
|
"ibm/granite-3-3-8b-instruct",
|
|
"ibm/granite-3-8b-instruct",
|
|
"ibm/granite-4-h-small",
|
|
"ibm/granite-8b-code-instruct",
|
|
|
|
# Llama Models
|
|
"meta-llama/llama-3-1-70b-gptq",
|
|
"meta-llama/llama-3-1-8b",
|
|
"meta-llama/llama-3-2-11b-vision-instruct",
|
|
"meta-llama/llama-3-2-90b-vision-instruct",
|
|
"meta-llama/llama-3-3-70b-instruct",
|
|
"meta-llama/llama-3-405b-instruct",
|
|
"meta-llama/llama-4-maverick-17b-128e-instruct-fp8",
|
|
|
|
# Mistral Models
|
|
"mistral-large-2512",
|
|
"mistralai/mistral-medium-2505",
|
|
"mistralai/mistral-small-3-1-24b-instruct-2503",
|
|
|
|
# Other Models
|
|
"openai/gpt-oss-120b",
|
|
|
|
# Embedding Models
|
|
"ibm/slate-125m-english-rtrvr",
|
|
"ibm/slate-30m-english-rtrvr",
|
|
]
|
|
|
|
|
|
@router.get(
|
|
"/v1/models",
|
|
response_model=ModelsResponse,
|
|
)
|
|
async def list_models():
|
|
"""List available models in OpenAI-compatible format.
|
|
|
|
Returns a list of models that can be used with the API.
|
|
Includes both the actual watsonx model IDs and any mapped names.
|
|
"""
|
|
created_time = int(time.time())
|
|
models = []
|
|
|
|
# Add all available watsonx models
|
|
for model_id in AVAILABLE_MODELS:
|
|
models.append(
|
|
ModelInfo(
|
|
id=model_id,
|
|
created=created_time,
|
|
owned_by="ibm-watsonx",
|
|
)
|
|
)
|
|
|
|
# Add mapped model names (e.g., gpt-4 -> ibm/granite-4-h-small)
|
|
model_mapping = settings.get_model_mapping()
|
|
for openai_name, watsonx_id in model_mapping.items():
|
|
if watsonx_id in AVAILABLE_MODELS:
|
|
models.append(
|
|
ModelInfo(
|
|
id=openai_name,
|
|
created=created_time,
|
|
owned_by="ibm-watsonx",
|
|
)
|
|
)
|
|
|
|
return ModelsResponse(data=models)
|
|
|
|
|
|
@router.get(
|
|
"/v1/models/{model_id}",
|
|
response_model=ModelInfo,
|
|
)
|
|
async def retrieve_model(model_id: str):
|
|
"""Retrieve information about a specific model.
|
|
|
|
Args:
|
|
model_id: The model ID to retrieve
|
|
|
|
Returns:
|
|
Model information
|
|
"""
|
|
# Map the model if needed
|
|
watsonx_model = settings.map_model(model_id)
|
|
|
|
# Check if model exists
|
|
if watsonx_model not in AVAILABLE_MODELS:
|
|
from fastapi import HTTPException
|
|
raise HTTPException(
|
|
status_code=404,
|
|
detail={
|
|
"error": {
|
|
"message": f"Model '{model_id}' not found",
|
|
"type": "invalid_request_error",
|
|
"code": "model_not_found",
|
|
}
|
|
},
|
|
)
|
|
|
|
return ModelInfo(
|
|
id=model_id,
|
|
created=int(time.time()),
|
|
owned_by="ibm-watsonx",
|
|
)
|