llama-stack/llama_stack/providers/utils/inference/__init__.py
Ashwin Bharambe e84d4436b5
Since we are pushing for HF repos, we should accept them in inference configs (#497)
# What does this PR do?

As the title says. 

## Test Plan

This needs
8752149f58
to also land. So the next package (0.0.54) will make this work properly.

The test is:

```bash
pytest -v -s -m "llama_3b and meta_reference" test_model_registration.py
```
2024-11-20 16:14:37 -08:00

38 lines
1 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 typing import List
from llama_models.datatypes import * # noqa: F403
from llama_models.sku_list import all_registered_models
def is_supported_safety_model(model: Model) -> bool:
if model.quantization_format != CheckpointQuantizationFormat.bf16:
return False
model_id = model.core_model_id
return model_id in [
CoreModelId.llama_guard_3_8b,
CoreModelId.llama_guard_3_1b,
CoreModelId.llama_guard_3_11b_vision,
]
def supported_inference_models() -> List[Model]:
return [
m
for m in all_registered_models()
if (
m.model_family in {ModelFamily.llama3_1, ModelFamily.llama3_2}
or is_supported_safety_model(m)
)
]
ALL_HUGGINGFACE_REPOS_TO_MODEL_DESCRIPTOR = {
m.huggingface_repo: m.descriptor() for m in all_registered_models()
}