Local LLM model page

Granite 3.2 Vision (2B)

IBM vision model for document extraction. Tiny but effective at understanding documents. 365K downloads.

Parameters
2B
Minimum RAM
4 GB
Model size
1.4 GB
Quantization
Q5_K_M

Can Granite 3.2 Vision (2B) run locally?

Granite 3.2 Vision (2B) is best suited for entry-level laptops and desktops. LocalClaw recommends Q5_K_M as the default quantization, with at least 4 GB RAM.

Search term for LM Studio or compatible runtimes: granite-3.2-2b-vision

Hugging Face repository: ibm-granite/granite-3.2-2b-vision-GGUF

visionlightspeed

Strengths

  • IBM vision model for document extraction. Tiny but effective at understanding documents. 365K downloads.

Limitations

  • Performance depends heavily on quantization, RAM bandwidth and runtime support.

Best use cases

  • vision
  • light
  • speed

Benchmarks

Speed: 10/10

Quality: 5/10

Coding: 3/10

Reasoning: 5/10

Technical details

Developer: granite

License: See model repository

Context window: Unknown tokens

Architecture: See model card

Released: 2025-02