Local LLM model page

Granite 3.3 (8B Instruct)

IBM enterprise-grade 8B. Trained for RAG, tool-use and structured output. Strong function calling and long-context performance (128K). Apache 2.0 with full data provenance.

Find the best model for my hardwareBrowse all 183 LLMs
Parameters
8B
Minimum RAM
8 GB
Model size
4.9 GB
Quantization
Q5_K_M

Can Granite 3.3 (8B Instruct) run locally?

Granite 3.3 (8B Instruct) is best suited for entry-level laptops and desktops. LocalClaw recommends Q5_K_M as the default quantization, with at least 8 GB RAM.

Search term for LM Studio or compatible runtimes: granite-3.3-8b-instruct

Hugging Face repository: ibm-granite/granite-3.3-8b-instruct-GGUF

chatcodestandardgeneralreasoning

Strengths

  • IBM enterprise-grade 8B. Trained for RAG, tool-use and structured output. Strong function calling and long-context performance (128K). Apache 2.0 with full data provenance.

Limitations

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

Best use cases

  • chat
  • code
  • standard
  • general
  • reasoning

Benchmarks

Speed: 8/10

Quality: 7/10

Coding: 7/10

Reasoning: 7/10

Technical details

Developer: granite

License: See model repository

Context window: Unknown tokens

Architecture: See model card

Released: 2025-10