Hugging Face's logo

bluetrace
/
SERA-32B-GA-FP8

Text Generation
Transformers
Safetensors
English
qwen3
fp8
quantized
llmcompressor
vllm
conversational
text-generation-inference
compressed-tensors
Model card Files
xet
 Community
ikarabulut-dev commited on
Commit
c5c9027
·
verified ·
1 Parent(s): 7bd2af6

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -26,7 +26,7 @@ FP8 quantization of [allenai/SERA-32B-GA](https://huggingface.co/allenai/SERA-32
26
  | Method | FP8 (W8A8) via `llmcompressor` `oneshot` |
27
  | Targets | All `Linear` layers except `lm_head` |
28
  | Calibration dataset | `allenai/Sera-4.5A-Lite-T2` |
29
- | Calibration samples | 513 |
30
  | Calibration sequence length | 2048 tokens |
31
  | llmcompressor version | 0.9.0.2 |
32
  | Hardware | AWS g6e.4xlarge (NVIDIA L40S, 48 GB VRAM) |
 
26
  | Method | FP8 (W8A8) via `llmcompressor` `oneshot` |
27
  | Targets | All `Linear` layers except `lm_head` |
28
  | Calibration dataset | `allenai/Sera-4.5A-Lite-T2` |
29
+ | Calibration samples | 512 |
30
  | Calibration sequence length | 2048 tokens |
31
  | llmcompressor version | 0.9.0.2 |
32
  | Hardware | AWS g6e.4xlarge (NVIDIA L40S, 48 GB VRAM) |