Report median instead of mean in bench_latency.py (#1269)
This commit is contained in:
@@ -233,7 +233,6 @@ python -m sglang.launch_server --model-path meta-llama/Meta-Llama-3-8B-Instruct
|
||||
### Supported Models
|
||||
|
||||
**Generative Models**
|
||||
- Exaone 3.0
|
||||
- Llama / Llama 2 / Llama 3 / Llama 3.1
|
||||
- Mistral / Mixtral / Mistral NeMo
|
||||
- Gemma / Gemma 2
|
||||
@@ -253,6 +252,7 @@ python -m sglang.launch_server --model-path meta-llama/Meta-Llama-3-8B-Instruct
|
||||
- Grok
|
||||
- ChatGLM
|
||||
- InternLM 2
|
||||
- Exaone 3
|
||||
|
||||
**Embedding Models**
|
||||
|
||||
|
||||
Reference in New Issue
Block a user