LLM EXPLORER 66,524 MODELS INDEXED

DeepSeek V4.1 Flash 1017 MLX Serve Lossless by ddalcu

By ddalcu · 859 downloads
HF Repo ↗ Copy URL

DeepSeek V4.1 Flash 1017 MLX Serve Lossless is an open-source language model by ddalcu. Features: 92.2b LLM, VRAM: 174.5GB, License: mit, LLM Explorer Score: 0.36.

Base model:deepseek-ai/deepsee... Base model:finetune:deepseek-a...   Deepseek v41   En   Engram   Mlx   Mlx-serve   Moe   Omlx   Region:us   Safetensors   Sharded   Tensorflow   Zh
Model Card on HF 🤗: https://huggingface.co/ddalcu/DeepSeek-V4.1-Flash-1017-MLX-Serve-Lossless 

DeepSeek V4.1 Flash 1017 MLX Serve Lossless Parameters and Internals

LLM NameDeepSeek V4.1 Flash 1017 MLX Serve Lossless
Repository 🤗https://huggingface.co/ddalcu/DeepSeek-V4.1-Flash-1017-MLX-Serve-Lossless 
Base Model(s)  deepseek-ai/DeepSeek-V4.1-Flash-1017   deepseek-ai/DeepSeek-V4.1-Flash-1017
Model Size92.2b
Required VRAM174.5 GB
Updated2026-11-03
Maintainerddalcu
Model Typedeepseek_v41
Model Files  3.6 GB: 1-of-80   5.0 GB: 2-of-80   2.6 GB: 3-of-80   5.0 GB: 4-of-80   4.8 GB: 5-of-80   2.6 GB: 6-of-80   4.8 GB: 7-of-80   2.6 GB: 8-of-80   4.8 GB: 9-of-80   3.9 GB: 10-of-80   4.8 GB: 11-of-80   2.6 GB: 12-of-80   4.8 GB: 13-of-80   2.6 GB: 14-of-80   4.8 GB: 15-of-80   2.6 GB: 16-of-80   4.8 GB: 17-of-80   2.6 GB: 18-of-80   4.8 GB: 19-of-80   2.6 GB: 20-of-80   4.8 GB: 21-of-80   2.6 GB: 22-of-80   4.8 GB: 23-of-80   2.6 GB: 24-of-80   4.8 GB: 25-of-80   2.6 GB: 26-of-80   4.8 GB: 27-of-80   2.6 GB: 28-of-80   4.8 GB: 29-of-80   4.1 GB: 30-of-80   4.8 GB: 31-of-80   2.6 GB: 32-of-80   4.8 GB: 33-of-80   2.6 GB: 34-of-80   4.8 GB: 35-of-80   2.6 GB: 36-of-80   4.8 GB: 37-of-80   2.6 GB: 38-of-80   4.8 GB: 39-of-80   4.2 GB: 40-of-80   3.5 GB: 41-of-80   5.0 GB: 42-of-80   4.8 GB: 43-of-80   2.6 GB: 44-of-80   4.8 GB: 45-of-80
Supported Languagesen zh
Model ArchitectureDeepseekV41ForCausalLM
Licensemit
Transformers Version5.6.0