Model comparison
Gemma 4 31B IT vs o3-mini
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 36.7 on the Noometry Index.
Last verified . 24 shared benchmarks.
GoogleSummary
- They share 24 benchmarks with published results for both. Gemma 4 31B IT scores higher in 7 categories and o3-mini in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 41.3 to 27.6.
- The biggest single-benchmark swing is LMCA: 39.3% for Gemma 4 31B IT and 19% for o3-mini.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- Gemma 4 31B IT accepts more context: 262K tokens versus 200K.
- Gemma 4 31B IT has downloadable open weights; the other is API-only.
Side by side
| Gemma 4 31B IT | o3-mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.2 | 36.7 |
| Released | 2026-04-02 | 2024-12-20 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 33K | 100K |
| Input $ / M tokens | $0.09 | $1.10 |
| Output $ / M tokens | $0.34 | $4.40 |
| Results tracked | 35 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemma 4 31B IT: 41.9 (#98), o3-mini: 41.4 (#116)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| SciCode | 43.4% | 39.8% |
| WeirdML | 52.3% | 43.7% |
| LMArena Coding | 1459 | 1378 |
| Aider Polyglot | — | 60.4% |
| LMArena WebDev | 1364 | — |
| GSO | — | 1.3% |
| LiveBench Coding | — | 82.7% |
| CadEval | — | 54% |
| ALE-Bench | 925.5 | — |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, o3-mini: 29.3 (#79)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| Cybench | — | 22.5% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 26.4 (#119), o3-mini: 16.3 (#295)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| CritPt | 1.4% | 0.3% |
| Chess Puzzles | 5% | 17% |
| LMArena Hard Prompts | 1448 | 1366 |
| DTBench | 82.7% | 68.8% |
| LMCA | 39.3% | 19% |
| Epoch Capabilities Index | 142.74 | 140.35 |
| ARC-AGI-2 | — | 3% |
| SimpleBench | — | 22.8% |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| ARC-AGI-1 | — | 34.5% |
| Thematic Generalization | 53% | — |
| LiveBench Reasoning | — | 89.6% |
| Mystery Game Puzzles | — | 7% |
| LiveBench Data Analysis | — | 70.6% |
| Surface Evolver Bench | 30.6% | — |
| ForecastBench | — | 59.6 |
| LiveBench | — | 75.9% |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 41.3 (#72), o3-mini: 27.6 (#236)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 76.9% |
| LMArena Math | 1464 | 1396 |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| LiveBench Math | — | 77.3% |
| MATH Level 5 | — | 96.5% |
| FrontierMath (Feb 2025 set) | — | 12.4% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Too close to call
Gemma 4 31B IT: 37.9 (#141), o3-mini: 38.3 (#135)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| GPQA Diamond | 75.8% | 77% |
| SimpleQA Verified | 10.4% | 15.3% |
| LMArena Expert | 1465 | 1364 |
| Confabulations | — | 17.9% |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Not comparable
Gemma 4 31B IT: 41.0 (#32), o3-mini: —
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| LMArena Vision | 1278 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 54.0 (#52), o3-mini: 45.9 (#159)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| LMArena Non-English | 1431 | 1319 |
| LMArena Chinese | 1478 | 1379 |
| LMArena French | 1436 | 1333 |
| LMArena Russian | 1462 | 1304 |
| LMArena Spanish | 1443 | 1321 |
| LMArena German | — | 1303 |
| LMArena Japanese | — | 1286 |
| LMArena Korean | — | 1314 |
Instruction Following Too close to call
Gemma 4 31B IT: 75.6 (#58), o3-mini: 75.3 (#67)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1433 | 1338 |
| LiveBench Instruction Following | — | 84.4% |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#67), o3-mini: 33.8 (#251)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| LMArena Longer Query | 1446 | 1343 |
| Fiction.LiveBench | — | 50% |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.7 (#92), o3-mini: 50.4 (#177)
| Benchmark | Gemma 4 31B IT | o3-mini |
|---|---|---|
| LMArena Text | 1443 | 1337 |
| LMArena Creative Writing | 1416 | 1286 |
| LMArena Multi-Turn | 1452 | 1320 |
| Short-Story Creative Writing | — | 61.7% |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
| LiveBench Language | — | 50.7% |
Frequently asked questions
Is Gemma 4 31B IT better than o3-mini?
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 36.7 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or o3-mini?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is Gemma 4 31B IT or o3-mini better for coding?
They score almost the same on coding (41.9 vs 41.4); test both on your own repository before choosing.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 200K.
How many benchmarks do Gemma 4 31B IT and o3-mini share?
24 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and o3-mini has 51.