llama-stack-mirror/llama_stack/templates/open-benchmark/build.yaml
Botao Chen 4dccf916d1
feat: open benchmark template and doc (#1465)
## What does this PR do?
- Provide a distro template to let developer easily run the open
benchmarks llama stack supports on llama and non-llama models.
- Provide doc on how to run open benchmark eval via CLI and open
benchmark contributing guide

[//]: # (If resolving an issue, uncomment and update the line below)
(Closes #1375 )

## Test Plan
open benchmark eval results on llama, gpt, gemini and clause
<img width="771" alt="Screenshot 2025-03-06 at 7 33 05 PM"
src="https://github.com/user-attachments/assets/1bd85456-b9b9-4b37-af76-4ce1d2bac00e"
/>

doc preview
<img width="944" alt="Screenshot 2025-03-06 at 7 33 58 PM"
src="https://github.com/user-attachments/assets/f4e5866d-b395-4c40-aa8b-080edeb5cdb6"
/>
<img width="955" alt="Screenshot 2025-03-06 at 7 34 04 PM"
src="https://github.com/user-attachments/assets/629defb6-d5e4-473c-aa03-308bce386fb4"
/>

<img width="965" alt="Screenshot 2025-03-06 at 7 35 29 PM"
src="https://github.com/user-attachments/assets/c21ff96c-9e8c-4c54-b6b8-25883125f4cf"
/>

<img width="957" alt="Screenshot 2025-03-06 at 7 35 37 PM"
src="https://github.com/user-attachments/assets/47571c90-1381-4e2c-bbed-c4f3a60578d0"
/>
2025-03-07 10:37:55 -08:00

36 lines
805 B
YAML

version: '2'
distribution_spec:
description: Distribution for running open benchmarks
providers:
inference:
- remote::openai
- remote::anthropic
- remote::gemini
- remote::groq
- remote::together
vector_io:
- inline::sqlite-vec
- remote::chromadb
- remote::pgvector
safety:
- inline::llama-guard
agents:
- inline::meta-reference
telemetry:
- inline::meta-reference
eval:
- inline::meta-reference
datasetio:
- remote::huggingface
- inline::localfs
scoring:
- inline::basic
- inline::llm-as-judge
- inline::braintrust
tool_runtime:
- remote::brave-search
- remote::tavily-search
- inline::code-interpreter
- inline::rag-runtime
- remote::model-context-protocol
image_type: conda