Model comparison
Granite 4.1 8b vs o3-pro
o3-pro is the stronger model overall, scoring 42.7 to 36.8 on the Noometry Index.
Last verified . 0 shared benchmarks.
Summary
- The widest gap is in long context, where o3-pro leads 72.1 to 38.7.
- Granite 4.1 8b has downloadable open weights; the other is API-only.
Side by side
| Granite 4.1 8b | o3-pro | |
|---|---|---|
| Provider | IBM | OpenAI |
| Noometry Index | 36.8 | 42.7 |
| Released | — | 2025-06-10 |
| Weights | Open | Proprietary |
| Context window | — | 200K |
| Max output | — | 100K |
| Input $ / M tokens | — | $20 |
| Output $ / M tokens | — | $80 |
| Results tracked | 13 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
Granite 4.1 8b: 29.9 (#291), o3-pro: 55.1 (#22)
| Benchmark | Granite 4.1 8b | o3-pro |
|---|---|---|
| Aider Polyglot | — | 84.9% |
| LMArena WebDev | 1190 | — |
| WeirdML | — | 58.2% |
| LMArena Coding | 1313 | — |
Reasoning Too close to call
Granite 4.1 8b: 24.6 (#142), o3-pro: 24.1 (#152)
| Benchmark | Granite 4.1 8b | o3-pro |
|---|---|---|
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 72.1% |
| ARC-AGI-1 | — | 59.3% |
| LMArena Hard Prompts | 1294 | — |
| DTBench | — | 86.9% |
| LMCA | — | 38.5% |
| Epoch Capabilities Index | — | 147.43 |
Math Not comparable
Granite 4.1 8b: 33.8 (#167), o3-pro: —
| Benchmark | Granite 4.1 8b | o3-pro |
|---|---|---|
| LMArena Math | 1311 | — |
Knowledge Granite 4.1 8b leads
Granite 4.1 8b: 35.5 (#171), o3-pro: 29.0 (#234)
| Benchmark | Granite 4.1 8b | o3-pro |
|---|---|---|
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 23.3% |
| LMArena Expert | 1310 | — |
Multilingual Not comparable
Granite 4.1 8b: 41.8 (#199), o3-pro: —
| Benchmark | Granite 4.1 8b | o3-pro |
|---|---|---|
| LMArena Non-English | 1261 | — |
| LMArena Chinese | 1335 | — |
| LMArena Russian | 1238 | — |
Instruction Following Not comparable
Granite 4.1 8b: 67.0 (#198), o3-pro: —
| Benchmark | Granite 4.1 8b | o3-pro |
|---|---|---|
| LMArena Instruction Following | 1268 | — |
Long Context o3-pro leads
Granite 4.1 8b: 38.7 (#188), o3-pro: 72.1 (#1)
| Benchmark | Granite 4.1 8b | o3-pro |
|---|---|---|
| Fiction.LiveBench | — | 97.2% |
| LMArena Longer Query | 1275 | — |
Writing & Preference o3-pro leads
Granite 4.1 8b: 48.3 (#199), o3-pro: 57.1 (#128)
| Benchmark | Granite 4.1 8b | o3-pro |
|---|---|---|
| LMArena Text | 1291 | — |
| LMArena Creative Writing | 1249 | — |
| Short-Story Creative Writing | — | 84.4% |
| LMArena Multi-Turn | 1269 | — |
Frequently asked questions
Is Granite 4.1 8b better than o3-pro?
o3-pro is the stronger model overall, scoring 42.7 to 36.8 on the Noometry Index.
Is Granite 4.1 8b or o3-pro better for coding?
o3-pro scores higher on coding benchmarks: 55.1 versus 29.9 in the Noometry coding category.
How many benchmarks do Granite 4.1 8b and o3-pro share?
0 benchmarks have published results for both models. Granite 4.1 8b has 13 scored results on Noometry and o3-pro has 12.