Update README.md
Browse files
README.md
CHANGED
|
@@ -112,7 +112,7 @@ The model was evaluated on AIME24, GPQA Diamond, MATH-500, and GSM8K benchmarks.
|
|
| 112 |
### Reproduction
|
| 113 |
|
| 114 |
The results can be reproduced using vLLM emulation docker: `rocmshared/pytorch:vllm-gfx950-mxfp4-mxfp6-v3`.
|
| 115 |
-
The results of AIME24, MATH-500, and GPQA Diamond, were obtained using [vLLM](https://docs.vllm.ai/en/latest/) while GSM8K was obtained using [SGLang](https://docs.sglang.ai/). For AIME24, MATH-500, and GPQA Diamond, we took 10 rounds with different random seeds for reliable performance
|
| 116 |
|
| 117 |
```
|
| 118 |
MODEL_ARGS="model_name=amd/DeepSeek-R1-0528-MXFP4-ASQ,dtype=bfloat16,tensor_parallel_size=8,max_model_length=71536,max_num_batched_tokens=32768,gpu_memory_utilization=0.85,generation_parameters={max_new_tokens:65536,temperature:0.6,top_p:0.95,seed:$SEED}"
|
|
|
|
| 112 |
### Reproduction
|
| 113 |
|
| 114 |
The results can be reproduced using vLLM emulation docker: `rocmshared/pytorch:vllm-gfx950-mxfp4-mxfp6-v3`.
|
| 115 |
+
The results of AIME24, MATH-500, and GPQA Diamond, were obtained using [vLLM](https://docs.vllm.ai/en/latest/) while GSM8K was obtained using [SGLang](https://docs.sglang.ai/). For AIME24, MATH-500, and GPQA Diamond, we took 10 rounds with different random seeds for reliable performance.
|
| 116 |
|
| 117 |
```
|
| 118 |
MODEL_ARGS="model_name=amd/DeepSeek-R1-0528-MXFP4-ASQ,dtype=bfloat16,tensor_parallel_size=8,max_model_length=71536,max_num_batched_tokens=32768,gpu_memory_utilization=0.85,generation_parameters={max_new_tokens:65536,temperature:0.6,top_p:0.95,seed:$SEED}"
|