LLM EXPLORER 55,709 MODELS INDEXED

Nanonets OCR S Bf16 by mlx-community

By mlx-community · 96 downloads
HF Repo ↗ Copy URL

Nanonets OCR S Bf16 is an open-source language model by mlx-community. Features: 3b LLM, VRAM: 7.5GB, Context: 125K, Instruction-Based.

Base model:finetune:qwen/qwen2... Base model:qwen/qwen2.5-vl-3b-...   Conversational   En   Endpoints compatible   Image-text-to-text   Instruct   Mlx   Ocr   Pdf2markdown   Qwen2 5 vl   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/mlx-community/Nanonets-OCR-s-bf16 

Nanonets OCR S Bf16 Parameters and Internals

LLM NameNanonets OCR S Bf16
Repository 🤗https://huggingface.co/mlx-community/Nanonets-OCR-s-bf16 
Base Model(s)  Qwen/Qwen2.5-VL-3B-Instruct   Qwen/Qwen2.5-VL-3B-Instruct
Model Size3b
Required VRAM7.5 GB
Updated2026-08-08
Maintainermlx-community
Model Typeqwen2_5_vl
Instruction-BasedYes
Model Files  5.4 GB: 1-of-2   2.1 GB: 2-of-2
Supported Languagesen
Model ArchitectureQwen2_5_VLForConditionalGeneration
Context Length128000
Model Max Length128000
Transformers Version4.52.3
Padding Token<|endoftext|>
Vocabulary Size151936
Errorsreplace

Best Alternatives to Nanonets OCR S Bf16

Best Alternatives
Context / RAM
Downloads
Likes
Nanonets OCR2 3B125K / 7.5 GB659329511
Qwen2.5 VL 3B Instruct125K / 7.5 GB27106
...2.5 VL 3B Instruct Abliterated125K / 7.5 GB501
Qwen2.5 VL 3B Instruct Bf16125K / 7.5 GB1554
Holo1 3B 3bit125K / 2.7 GB211
Qwen2.5 VL 3B Instruct 3bit125K / 2.7 GB2212
Holo1 3B 4bit125K / 3.1 GB271
Qwen2.5 VL 3B Instruct 4bit125K / 3.1 GB32355
Qwen2.5 VL 3B Instruct 6bit125K / 3.2 GB983
Qwen2.5 VL 3B Instruct 8bit125K / 4.1 GB5287
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/Nanonets-OCR-s-bf16.