Hugging Face's logo

RedHatAI
/
Kimi-K3-NVFP4

Feature Extraction
Transformers
Safetensors
kimi_k3
compressed-tensors
LLM Compressor
vLLM
custom_code
8-bit precision
Model card Files
xet
 Community
kylesayrs commited on
Commit
71eed69
·
verified ·
1 Parent(s): 5b6b06a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -32,9 +32,9 @@ May require https://github.com/vllm-project/vllm/pull/50500 to run
32
  This model was compressed using LLM Compressor. For more information, please contact Kyle Sayers via the [vLLM Developers Slack](https://communityinviter.com/apps/vllm-dev/join-vllm-developers-slack) or ksayers@redhat.com.
33
 
34
  ## Evaluation ##
35
- | Benchmark | `deepseek-ai/DeepSeek-V4-Pro-Base` | `deepseek-ai/DeepSeek-V4-Pro` | `RedHatAI/DeepSeek-V4-Pro-NVFP4-FP8` |
36
- | - | - | -| - |
37
- | GPQA | | 93.5 | 91.0 |
38
 
39
  ```bash
40
  inspect eval hf/Idavidrein/gpqa/diamond \
 
32
  This model was compressed using LLM Compressor. For more information, please contact Kyle Sayers via the [vLLM Developers Slack](https://communityinviter.com/apps/vllm-dev/join-vllm-developers-slack) or ksayers@redhat.com.
33
 
34
  ## Evaluation ##
35
+ | Benchmark | `moonshotai/Kimi-K3` | `RedHatAI/Kimi-K3-NVFP4`
36
+ | - | - | -|
37
+ | GPQA | 93.5 | 91.0 |
38
 
39
  ```bash
40
  inspect eval hf/Idavidrein/gpqa/diamond \