LLM EXPLORER 57,918 MODELS INDEXED

Kimi Linear 48B A3B Instruct 8bit by mlx-community

By mlx-community · 204 downloads
HF Repo ↗ Copy URL

Kimi Linear 48B A3B Instruct 8bit is an open-source language model by mlx-community. Features: 48b LLM, VRAM: 52.2GB, License: mit, Quantized, Instruction-Based, LLM Explorer Score: 0.21.

  8-bit   8bit Base model:moonshotai/kimi-lin... Base model:quantized:moonshota...   Conversational   Custom code   Instruct   Kimi linear   Mlx   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/mlx-community/Kimi-Linear-48B-A3B-Instruct-8bit 

Kimi Linear 48B A3B Instruct 8bit Parameters and Internals

LLM NameKimi Linear 48B A3B Instruct 8bit
Repository 🤗https://huggingface.co/mlx-community/Kimi-Linear-48B-A3B-Instruct-8bit 
Base Model(s)  Kimi Linear 48B A3B Instruct   moonshotai/Kimi-Linear-48B-A3B-Instruct
Model Size48b
Required VRAM52.2 GB
Updated2026-08-10
Maintainermlx-community
Model Typekimi_linear
Instruction-BasedYes
Model Files  5.1 GB: 1-of-10   5.3 GB: 2-of-10   5.2 GB: 3-of-10   5.3 GB: 4-of-10   5.3 GB: 5-of-10   5.2 GB: 6-of-10   5.3 GB: 7-of-10   5.3 GB: 8-of-10   5.2 GB: 9-of-10   5.0 GB: 10-of-10
Quantization Type8bit
Model ArchitectureKimiLinearForCausalLM
Licensemit
Model Max Length1048576
Transformers Version4.57.1
Tokenizer ClassTikTokenTokenizer
Padding Token[PAD]
Vocabulary Size163840

Best Alternatives to Kimi Linear 48B A3B Instruct 8bit

Best Alternatives
Context / RAM
Downloads
Likes
...i Linear 48B A3B Instruct 4bit0K / 27.6 GB11519
...i Linear 48B A3B Instruct 6bit0K / 39.8 GB3903
...i Linear 48B A3B Instruct 3bit0K / 21.6 GB1432
...i Linear 48B A3B Instruct 5bit0K / 33.7 GB1121
Kimi Linear 48B A3B Instruct0K / 98.3 GB71928566
...i Linear REAP 35B A3B Instruct0K / 70.3 GB6068
...near 48B A3B Instruct Mlx Bf160K / 98 GB1081
...r 48B A3B Instruct Abliterated0K / 98.3 GB1229
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/Kimi-Linear-48B-A3B-Instruct-8bit.