LLM EXPLORER 65,418 MODELS INDEXED

DeepSeek V4.1 Flash 1017 MLX Serve IQ MLX 3.0bpw by ddalcu

By ddalcu · 624 downloads
HF Repo ↗ Copy URL

DeepSeek V4.1 Flash 1017 MLX Serve IQ MLX 3.0bpw is an open-source language model by ddalcu. Features: 66.2b LLM, VRAM: 222.4GB, License: mit, LLM Explorer Score: 0.33.

Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-...   Deepseek v41   Dspark   En   Engram   Gptq   Mlx   Mlx-serve   Moe   Region:us   Safetensors   Sharded   Tensorflow   Zh
Model Card on HF 🤗: https://huggingface.co/ddalcu/DeepSeek-V4.1-Flash-1017-MLX-Serve-iQ-MLX-3.0bpw 

DeepSeek V4.1 Flash 1017 MLX Serve IQ MLX 3.0bpw Parameters and Internals

LLM NameDeepSeek V4.1 Flash 1017 MLX Serve IQ MLX 3.0bpw
Repository 🤗https://huggingface.co/ddalcu/DeepSeek-V4.1-Flash-1017-MLX-Serve-iQ-MLX-3.0bpw 
Base Model(s)  deepseek-ai/DeepSeek-V4.1-Flash-1017   deepseek-ai/DeepSeek-V4.1-Flash-1017
Model Size66.2b
Required VRAM222.4 GB
Updated2026-10-22
Maintainerddalcu
Model Typedeepseek_v41
Model Files  6.1 GB: 1-of-36   5.5 GB: 2-of-36   5.4 GB: 3-of-36   6.2 GB: 4-of-36   6.0 GB: 5-of-36   6.0 GB: 6-of-36   6.0 GB: 7-of-36   6.0 GB: 8-of-36   6.0 GB: 9-of-36   6.0 GB: 10-of-36   6.0 GB: 11-of-36   6.0 GB: 12-of-36   6.0 GB: 13-of-36   6.2 GB: 14-of-36   6.2 GB: 15-of-36   6.8 GB: 16-of-36   6.0 GB: 17-of-36   6.2 GB: 18-of-36   6.2 GB: 19-of-36   6.8 GB: 20-of-36   6.2 GB: 21-of-36   6.2 GB: 22-of-36   6.8 GB: 23-of-36   6.2 GB: 24-of-36   6.2 GB: 25-of-36   6.8 GB: 26-of-36   6.2 GB: 27-of-36   6.2 GB: 28-of-36   6.8 GB: 29-of-36   6.2 GB: 30-of-36   6.2 GB: 31-of-36   6.8 GB: 32-of-36   6.2 GB: 33-of-36   6.2 GB: 34-of-36   6.8 GB: 35-of-36   4.8 GB: 36-of-36
Supported Languagesen zh
Model ArchitectureDeepseekV41ForCausalLM
Licensemit
Model Max Length1048576
Transformers Version5.6.0
Tokenizer ClassPreTrainedTokenizerFast
Beginning of Sentence Token<｜begin▁of▁sentence｜>
End of Sentence Token<｜end▁of▁sentence｜>