What are the hardware requirements for DeepSeek OCR 2 Bf16?

DeepSeek OCR 2 Bf16 requires approximately 6.7 GB of VRAM and supports a context window of 8K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed DeepSeek OCR 2 Bf16 and how large is it?

DeepSeek OCR 2 Bf16 is developed by mlx-community, a model with 3.4b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate DeepSeek OCR 2 Bf16?

DeepSeek OCR 2 Bf16 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

DeepSeek OCR 2 Bf16 by mlx-community — VRAM 6.7GB, 8K context

Name: DeepSeek OCR 2 Bf16
Author: mlx-community

DeepSeek OCR 2 Bf16 is an open-source language model by mlx-community. Features: 3.4b LLM, VRAM: 6.7GB, Context: 8K, License: apache-2.0, LLM Explorer Score: 0.24.

Conversational Custom code Deepseek Deepseekocr 2 Feature-extraction Image-text-to-text Mlx Multilingual Ocr Region:us Safetensors Sharded Tensorflow Vision Vision-language

Model Card on HF 🤗: https://huggingface.co/mlx-community/DeepSeek-OCR-2-bf16

DeepSeek OCR 2 Bf16 Parameters and Internals

LLM Name	DeepSeek OCR 2 Bf16
Repository 🤗	https://huggingface.co/mlx-community/DeepSeek-OCR-2-bf16
Model Size	3.4b
Required VRAM	6.7 GB
Updated	2026-08-08
Maintainer	mlx-community
Model Type	deepseekocr_2
Model Files	5.3 GB: 1-of-2 1.4 GB: 2-of-2
Model Architecture	DeepseekOCR2ForCausalLM
License	apache-2.0
Context Length	8192
Model Max Length	8192
Transformers Version	4.46.3
Tokenizer Class	TokenizersBackend
Padding Token	<｜▁pad▁｜>
Vocabulary Size	129280

Rank the DeepSeek OCR 2 Bf16 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to DeepSeek OCR 2 Bf16

Best Alternatives	Context / RAM	Downloads	Likes
DeepSeek OCR 2	8K / 6.8 GB	2145486	1068
DeepSeek OCR 2	8K / 6.8 GB	2348	53

Note: green Score (e.g. "73.2") means that the model is better than mlx-community/DeepSeek-OCR-2-bf16.