LLM EXPLORER 61,652 MODELS INDEXED

LFM2.5 2.6B OptiQ 4bit by mlx-community

By mlx-community · 337 downloads
HF Repo ↗ Copy URL

LFM2.5 2.6B OptiQ 4bit is an open-source language model by mlx-community. Features: 2.6b LLM, VRAM: 2GB, Context: 128K, License: other, Quantized, LLM Explorer Score: 0.27.

  4-bit   4bit   Apple-silicon Base model:liquidai/lfm2.5-2.6... Base model:quantized:liquidai/...   Conversational   Lfm2   Liquid-ai   Mixed-precision   Mlx   On-device   Optiq   Quantized   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/mlx-community/LFM2.5-2.6B-OptiQ-4bit 

LFM2.5 2.6B OptiQ 4bit Parameters and Internals

LLM NameLFM2.5 2.6B OptiQ 4bit
Repository 🤗https://huggingface.co/mlx-community/LFM2.5-2.6B-OptiQ-4bit 
Base Model(s)  LiquidAI/LFM2.5-2.6B   LiquidAI/LFM2.5-2.6B
Model Size2.6b
Required VRAM2 GB
Updated2026-09-19
Maintainermlx-community
Model Typelfm2
Model Files  2.0 GB
Quantization Type4bit
Model ArchitectureLfm2ForCausalLM
Licenseother
Context Length131072
Model Max Length131072
Transformers Version5.2.0
Tokenizer ClassTokenizersBackend
Padding Token<|pad|>
Vocabulary Size128000

Best Alternatives to LFM2.5 2.6B OptiQ 4bit

Best Alternatives
Context / RAM
Downloads
Likes
LFM2.5 2.6B MLX 8bit128K / 2.9 GB601
LFM2.5 2.6B MLX 6bit128K / 2.2 GB521
LFM2.5 2.6B MLX 5bit128K / 1.9 GB491
LFM2.5 2.6B MLX 4bit128K / 1.6 GB421
LFM2.5 2.6B 4bit125K / 1.5 GB15093
LFM2.5 2.6B 8bit125K / 2.9 GB8543
LFM2 2.6B Exp 4bit125K / 1.4 GB1801
LFM2 2.6B Transcript 4bit125K / 1.4 GB481
LFM2 2.6B Exp 8bit125K / 2.7 GB162
LFM2 2.6B Exp 6bit125K / 2.1 GB141
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/LFM2.5-2.6B-OptiQ-4bit.