benchmarks
Granite 4.0 350M is a compact, 350-million-parameter language model from IBM's Granite family. Its downloadable weights support local text-generation experiments and customization; this profile applies specifically to the 350M model.
Benchmark results
Every figure links to where it was published. Compare like with like — harnesses and prompt setups differ between vendors.
| Benchmark | Score | Source |
|---|---|---|
| GPQA DiamondPhD-level science questions | 11.2%28 Aug 2026 | View |
| OTIS Mock AIME 2024-25Competition mathematics | 0.3%28 Aug 2026 | View |
Weighted mean of 2 verified benchmark results: GPQA Diamond 11.2%, OTIS Mock AIME 2024-25 0.3%.
Capabilities
- Text generation
- Compact 350-million-parameter model size
- Downloadable weights for local deployment
- Customization through fine-tuning
What improved
- Not yet confirmed by public reporting.
Known flaws & limitations
- Not yet confirmed by public reporting.
Risks, safety & security issues
- Not yet confirmed by public reporting.
Good fit for
- Local text-generation prototyping
- Small-model research and experimentation
- Task-specific fine-tuning with independent evaluation
Poor fit for
- Deployments requiring established performance on the listed reasoning and agent benchmarks
- High-stakes decisions without independent validation and human oversight
Profile last reviewed 10/20/2026 and re-checked automatically every few hours against the vendor's published documentation, papers and public leaderboards. Scores come from the cited benchmark results above — confirm limits and pricing against the provider's own documentation before building on them.