phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-31 07:33:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge-related changes.

Browse source
This commit is contained in:
ilya-kolchinsky 2025-04-02 19:56:44 +02:00
commit 60e9f46856
456 changed files with 38636 additions and 10892 deletions
Download patch file Download diff file Expand all files Collapse all files

6
llama_stack/templates/tgi/doc_template.md
View file

@ -38,6 +38,7 @@ export INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct
export CUDA_VISIBLE_DEVICES=0
docker run --rm -it \
--pull always \
-v $HOME/.cache/huggingface:/data \
-p $INFERENCE_PORT:$INFERENCE_PORT \
--gpus $CUDA_VISIBLE_DEVICES \
@ -58,6 +59,7 @@ export SAFETY_MODEL=meta-llama/Llama-Guard-3-1B
export CUDA_VISIBLE_DEVICES=1
docker run --rm -it \
--pull always \
-v $HOME/.cache/huggingface:/data \
-p $SAFETY_PORT:$SAFETY_PORT \
--gpus $CUDA_VISIBLE_DEVICES \
@ -78,9 +80,10 @@ Now you are ready to run Llama Stack with TGI as the inference provider. You can
This method allows you to get started quickly without having to build the distribution code.
```bash
LLAMA_STACK_PORT=5001
LLAMA_STACK_PORT=8321
docker run \
-it \
--pull always \
-p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
llamastack/distribution-{{ name }} \
--port $LLAMA_STACK_PORT \
@ -97,6 +100,7 @@ cd /path/to/llama-stack
docker run \
-it \
--pull always \
-p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
-v ~/.llama:/root/.llama \
-v ./llama_stack/templates/tgi/run-with-safety.yaml:/root/my-run.yaml \

23
llama_stack/templates/tgi/run-with-safety.yaml
View file

@ -32,7 +32,8 @@ providers:
safety:
- provider_id: llama-guard
provider_type: inline::llama-guard
config: {}
config:
excluded_categories: []
agents:
- provider_id: meta-reference
provider_type: inline::meta-reference
@ -45,20 +46,32 @@ providers:
- provider_id: meta-reference
provider_type: inline::meta-reference
config:
service_name: ${env.OTEL_SERVICE_NAME:llama-stack}
service_name: "${env.OTEL_SERVICE_NAME:\u200B}"
sinks: ${env.TELEMETRY_SINKS:console,sqlite}
sqlite_db_path: ${env.SQLITE_DB_PATH:~/.llama/distributions/tgi/trace_store.db}
eval:
- provider_id: meta-reference
provider_type: inline::meta-reference
config: {}
config:
kvstore:
type: sqlite
namespace: null
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/meta_reference_eval.db
datasetio:
- provider_id: huggingface
provider_type: remote::huggingface
config: {}
config:
kvstore:
type: sqlite
namespace: null
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/huggingface_datasetio.db
- provider_id: localfs
provider_type: inline::localfs
config: {}
config:
kvstore:
type: sqlite
namespace: null
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/localfs_datasetio.db
scoring:
- provider_id: basic
provider_type: inline::basic

23
llama_stack/templates/tgi/run.yaml
View file

@ -31,7 +31,8 @@ providers:
safety:
- provider_id: llama-guard
provider_type: inline::llama-guard
config: {}
config:
excluded_categories: []
agents:
- provider_id: meta-reference
provider_type: inline::meta-reference
@ -44,20 +45,32 @@ providers:
- provider_id: meta-reference
provider_type: inline::meta-reference
config:
service_name: ${env.OTEL_SERVICE_NAME:llama-stack}
service_name: "${env.OTEL_SERVICE_NAME:\u200B}"
sinks: ${env.TELEMETRY_SINKS:console,sqlite}
sqlite_db_path: ${env.SQLITE_DB_PATH:~/.llama/distributions/tgi/trace_store.db}
eval:
- provider_id: meta-reference
provider_type: inline::meta-reference
config: {}
config:
kvstore:
type: sqlite
namespace: null
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/meta_reference_eval.db
datasetio:
- provider_id: huggingface
provider_type: remote::huggingface
config: {}
config:
kvstore:
type: sqlite
namespace: null
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/huggingface_datasetio.db
- provider_id: localfs
provider_type: inline::localfs
config: {}
config:
kvstore:
type: sqlite
namespace: null
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/localfs_datasetio.db
scoring:
- provider_id: basic
provider_type: inline::basic

2
llama_stack/templates/tgi/tgi.py
View file

@ -143,7 +143,7 @@ def get_distribution_template() -> DistributionTemplate:
},
run_config_env_vars={
"LLAMA_STACK_PORT": (
"5001",
"8321",
"Port for the Llama Stack distribution server",
),
"INFERENCE_MODEL": (