LLM EXPLORER 56,604 MODELS INDEXED

LFM2.5 1.2B Thinking MLX 4bit by LiquidAI

By LiquidAI · 343 downloads
HF Repo ↗ Copy URL

LFM2.5 1.2B Thinking MLX 4bit is an open-source language model by LiquidAI. Features: 1.2b LLM, VRAM: 0.7GB, Context: 125K, License: other, Quantized, LLM Explorer Score: 0.23.

  4-bit   4bit   Ar Base model:liquidai/lfm2.5-1.2... Base model:quantized:liquidai/...   Conversational   De   Edge   En   Es   Fr   It   Ja   Ko   Lfm2   Lfm2.5   Liquid   Mlx   Pt   Quantized   Reasoning   Region:us   Safetensors   Zh
Model Card on HF 🤗: https://huggingface.co/LiquidAI/LFM2.5-1.2B-Thinking-MLX-4bit 

LFM2.5 1.2B Thinking MLX 4bit Parameters and Internals

LLM NameLFM2.5 1.2B Thinking MLX 4bit
Repository 🤗https://huggingface.co/LiquidAI/LFM2.5-1.2B-Thinking-MLX-4bit 
Base Model(s)  LiquidAI/LFM2.5-1.2B-Thinking   LiquidAI/LFM2.5-1.2B-Thinking
Model Size1.2b
Required VRAM0.7 GB
Updated2026-08-08
MaintainerLiquidAI
Model Typelfm2
Model Files  0.7 GB
Supported Languagesen ja ko fr es de it pt ar zh
Quantization Type4bit
Model ArchitectureLfm2ForCausalLM
Licenseother
Context Length128000
Model Max Length128000
Transformers Version4.57.2
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|pad|>
Vocabulary Size65536

Best Alternatives to LFM2.5 1.2B Thinking MLX 4bit

Best Alternatives
Context / RAM
Downloads
Likes
LFM2.5 1.2B Instruct 4bit125K / 0.7 GB150151
LFM2.5 1.2B JP 202606 MLX 4bit125K / 0.7 GB2278
LFM2.5 1.2B JP 202606 MLX 8bit125K / 1.2 GB2267
LFM2.5 1.2B JP 202606 MLX 6bit125K / 1 GB2007
LFM2.5 1.2B JP 202606 MLX 5bit125K / 0.8 GB1888
LFM2.5 1.2B Instruct MLX 8bit125K / 1.2 GB68125
LFM2.5 1.2B Thinking MLX 8bit125K / 1.2 GB29415
LFM2.5 1.2B JP MLX 4bit125K / 0.7 GB20210
LFM2.5 1.2B JP MLX 8bit125K / 1.2 GB1189
LFM2.5 1.2B Thinking MLX 6bit125K / 1 GB925
Note: green Score (e.g. "73.2") means that the model is better than LiquidAI/LFM2.5-1.2B-Thinking-MLX-4bit.