Downloads · 30 days
24
34% of all-time downloads
Skyfall-31B-v4.2-NVFP4
vissarionovich/Skyfall-31B-v4.2-NVFP4
Skyfall-31B-v4.2-NVFP4 is a machine learning model from vissarionovich. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
NVFP4 quantization of TheDrummer/Skyfall-31B-v4.2
All-time downloads
70
Public
Parameters
31.4B
19.6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors19.6 GB · 100%
Parameter types
How the weights are stored.
U830B · 96%
Base models
Datasets
Model card
From the Hugging Face model README
Skyfall-31B-v4.2-NVFP4
NVFP4 quantization of TheDrummer/Skyfall-31B-v4.2
One-shot calibration with LLM Compressor, NVFP4 recipe:
| Calibration data | HuggingFaceH4/ultrachat_200k |
| Samples | 128 |
| Sequence length | 4096 |
| Targets | Linear |
| Ignored | lm_head |
recipe = QuantizationModifier(targets="Linear", scheme="NVFP4", ignore=["lm_head"])
if there's something catastrophically wrong relative to the original model, go complain on discussion tab, i'm mostly following a tutorial here.
also, haven't been tested yet, i'll edit this part out after.
- Type
- mistral
- Created
- Sep 17, 2026
- Updated
- Sep 17, 2026