phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

chore: remove straggler references to llama-models (#1345)

Browse source 
Straggler references cleanup
This commit is contained in:
Ashwin Bharambe 2025-03-01 14:26:03 -08:00 • committed by GitHub
parent 8bbd52bb9f
commit 46b0a404e8
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
19 changed files with 827 additions and 74 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama_stack/providers/inline/inference/meta_reference/inference.py
View file

@ -111,7 +111,7 @@ class MetaReferenceInferenceImpl(
)
if llama_model is None:
raise ValueError(
"Please make sure your llama_model in model metadata or model identifier is in llama-models SKU list"
"Please make sure your llama_model in model metadata or model identifier is in Llama SKU list"
)
self.model_registry_helper = ModelRegistryHelper(

2
llama_stack/providers/inline/inference/meta_reference/quantization/scripts/run_quantize_checkpoint.sh
View file

@ -21,7 +21,7 @@ NPROC=$7
echo $MASTER_HOST, $RUN_ID, $CKPT_DIR, $QUANT_CKPT_DIR
NCCL_NET=Socket NCCL_SOCKET_IFNAME=eth TIKTOKEN_CACHE_DIR="" PYTHONPATH="/home/$USER/llama-models:/home/$USER/llama-stack" \
NCCL_NET=Socket NCCL_SOCKET_IFNAME=eth TIKTOKEN_CACHE_DIR="" PYTHONPATH="/home/$USER/llama-stack" \
torchrun \
--nnodes=$NNODES --nproc_per_node=$NPROC \
--rdzv_id=$RUN_ID \

9
llama_stack/providers/remote/safety/bedrock/bedrock.py
View file

@ -59,7 +59,8 @@ class BedrockSafetyAdapter(Safety, ShieldsProtocolPrivate):
if not shield:
raise ValueError(f"Shield {shield_id} not found")
"""This is the implementation for the bedrock guardrails. The input to the guardrails is to be of this format
"""
This is the implementation for the bedrock guardrails. The input to the guardrails is to be of this format
```content = [
{
"text": {
@ -67,10 +68,8 @@ class BedrockSafetyAdapter(Safety, ShieldsProtocolPrivate):
}
}
]```
However the incoming messages are of this type UserMessage(content=....) coming from
https://github.com/meta-llama/llama-models/blob/main/models/llama3/api/datatypes.py
They contain content, role . For now we will extract the content and default the "qualifiers": ["query"]
Incoming messages contain content, role . For now we will extract the content and
default the "qualifiers": ["query"]
"""
shield_params = shield.params