hugpadhugpad

UraionLabs / NVIDIA-Nemotron-3.5-Bolt-E4B-MLX-oQ4e

Agents8.8B2571Updated Oct 23, 2026
Hugging Face

Downloads

257

Likes

1

Last updated

Oct 23, 2026

License

openmdw-1.1

NVIDIA Nemotron 3.5 Bolt E4B quantized for Apple Silicon with MLX: 5.6 GB, mixed precision, 4-bit base with attention held at 8-bit, imatrix-calibrated.

Tasks & library

  • text generation
  • mlx

Tags

  • mlx
  • gemma4
  • mlx-vlm
  • mlx-lm
  • apple-silicon
  • mac

Technical specifications

Parameters8.8B
Librarymlx
Languages34 languages
Base modelnvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16
Licenseopenmdw-1.1