LLM EXPLORER 68,292 MODELS INDEXED

DeepSeek V4 Flash 0731 MXFP4 MLX by TensorFold

By TensorFold · 1781 downloads

DeepSeek V4 Flash 0731 MXFP4 MLX is an open-source language model by TensorFold. Features: 304.2b LLM, VRAM: 2GB, Context: 1024K, License: mit, LLM Explorer Score: 0.28.

  4-bit Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-...   Deepseek   Deepseek v4   Mlx   Moe   Mxfp4   Mxfp8   Region:us   Safetensors

DeepSeek V4 Flash 0731 MXFP4 MLX Parameters and Internals

LLM NameDeepSeek V4 Flash 0731 MXFP4 MLX
Repository 🤗https://huggingface.co/TensorFold/DeepSeek-V4-Flash-0731-MXFP4-MLX 
Base Model(s)  DeepSeek V4 Flash 0731   deepseek-ai/DeepSeek-V4-Flash-0731
Model Size304.2b
Required VRAM2 GB
Updated2026-11-21
MaintainerTensorFold
Model Typedeepseek_v4
Model Files  5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   5.0 GB   2.0 GB
Model ArchitectureDeepseekV4ForCausalLM
Licensemit
Context Length1048576
Model Max Length1048576
Transformers Version4.57.1
Tokenizer ClassPreTrainedTokenizerFast
Beginning of Sentence Token<|begin▁of▁sentence|>
End of Sentence Token<|end▁of▁sentence|>
Vocabulary Size129280
Torch Data Typebfloat16

Best Alternatives to DeepSeek V4 Flash 0731 MXFP4 MLX

Best Alternatives
Context / RAM
Downloads
Likes
DeepSeek V4 Flash 07311024K / 152.3 GB7857712803
...Seek V4 Flash 0731 Abliterated1024K / 148.7 GB12646211
... V4 Flash 0731 Abliterated FP81024K / 145.1 GB514935
DeepSeek V4 Flash 0731 NVFP41024K / 149.3 GB74216
...4 Flash 0731 Abliterated NVFP41024K / 145.5 GB95211
DeepSeek V4 Flash 07311024K / 152.3 GB2531
DeepSeek V4 Flash 0731 NVFP41024K / 156.9 GB30723
... 0731 MXFP4 Mixed CT AutoRound1024K / 152.3 GB1541
...eepSeek V4 Flash 0731 Ornstein1024K / 568.7 GB1831
...Seek V4 Flash 0731 CRACK NVFP41024K / 141.5 GB2427
Note: green Score (e.g. "73.2") means that the model is better than TensorFold/DeepSeek-V4-Flash-0731-MXFP4-MLX.