LLM EXPLORER 61,060 MODELS INDEXED

DeepSeek V3.1 MLX 5.5bit by inferencerlabs

By inferencerlabs · 400 downloads
HF Repo ↗ Copy URL

DeepSeek V3.1 MLX 5.5bit is an open-source language model by inferencerlabs. Features: LLM, Context: 160K, License: mit, LLM Explorer Score: 0.2.

  5-bit Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-...   Conversational   Custom code   Deepseek v3   Mlx   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/inferencerlabs/DeepSeek-V3.1-MLX-5.5bit 

DeepSeek V3.1 MLX 5.5bit Parameters and Internals

LLM NameDeepSeek V3.1 MLX 5.5bit
Repository 🤗https://huggingface.co/inferencerlabs/DeepSeek-V3.1-MLX-5.5bit 
Base Model(s)  deepseek-ai/DeepSeek-V3.1   deepseek-ai/DeepSeek-V3.1
Updated2026-06-02
Maintainerinferencerlabs
Model Typedeepseek_v3
Model ArchitectureDeepseekV3ForCausalLM
Licensemit
Context Length163840
Model Max Length163840
Transformers Version4.44.2
Tokenizer ClassLlamaTokenizerFast
Padding Token<｜end▁of▁sentence｜>
Vocabulary Size129280
Torch Data Typebfloat16

Best Alternatives to DeepSeek V3.1 MLX 5.5bit

Best Alternatives
Context / RAM
Downloads
Likes
Kimi K2 Thinking NVFP4256K / 210 GB1556929
Kimi K2 Thinking BF16256K / 220 GB636
Kimi K2 Instruct 0905256K / 667.4 GB177
DeepSeek R1 0528160K / 176.1 GB4815
DeepSeek R1 0528 FP4160K / 107 GB7299136
DeepSeek V3.0324 FP4160K / 242.8 GB998412
DeepSeek R1 FP4160K / 232 GB6856264
...V3.1 Base Int4 Mixed AutoRound160K / 220 GB1032
...Seek V3.1 Int4 Mixed AutoRound160K / 225 GB5311
R1 1776160K / 243.4 GB229
Note: green Score (e.g. "73.2") means that the model is better than inferencerlabs/DeepSeek-V3.1-MLX-5.5bit.