Model comparison
Gemma 4 31B IT vs o4-mini
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 41.6 on the Noometry Index.
Last verified . 27 shared benchmarks.
GoogleSummary
- They share 27 benchmarks with published results for both. Gemma 4 31B IT scores higher in 7 categories and o4-mini in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemma 4 31B IT leads 53.8 to 47.0.
- The biggest single-benchmark swing is Chess Puzzles: 5% for Gemma 4 31B IT and 26% for o4-mini.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $1.10 / $4.40 for o4-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 | o4-mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 41.6 |
| Released | 2026-04-02 | 2025-04-16 |
| 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 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 4 31B IT: 42.3 (#108), o4-mini: 40.9 (#127)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| WeirdML | 52.3% | 52.6% |
| LMArena Coding | 1459 | 1368 |
| ALE-Bench | 925.5 | 826.17 |
| SWE-bench Verified (bash only) | — | 45% |
| Aider Polyglot | — | 72% |
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| GSO | — | 3.6% |
| CadEval | — | 62% |
| AlgoTune | — | 1.72 |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, o4-mini: 32.6 (#61)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| METR Time Horizons | — | 63.9% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), o4-mini: 24.6 (#162)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 67.6% |
| CritPt | 1.4% | 0.6% |
| Chess Puzzles | 5% | 26% |
| LMArena Hard Prompts | 1448 | 1351 |
| DTBench | 82.7% | 77.6% |
| LMCA | 39.3% | 26.5% |
| Epoch Capabilities Index | 142.74 | 145.64 |
| ARC-AGI-2 | — | 6.1% |
| SimpleBench | — | 38.7% |
| NYT Connections (extended) | 70.6% | — |
| ARC-AGI-1 | — | 58.7% |
| EnigmaEval | — | 9.2% |
| Thematic Generalization | 53% | — |
| Mystery Game Puzzles | — | 5% |
| Surface Evolver Bench | 30.6% | — |
| ForecastBench | — | 61.8 |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), o4-mini: 40.8 (#89)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 81.7% |
| LMArena Math | 1465 | 1389 |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| Omni-MATH | — | 72% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 24.8% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge o4-mini leads
Gemma 4 31B IT: 37.9 (#151), o4-mini: 43.6 (#91)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| GPQA Diamond | 75.8% | 79.6% |
| SimpleQA Verified | 10.4% | 19.6% |
| Vectara Hallucination Rate | 7.4% | 18.6% |
| LMArena Expert | 1465 | 1343 |
| Humanity's Last Exam | — | 18.1% |
| MMLU-Pro | — | 82% |
| Confabulations | — | 15.8% |
| GPQA (HELM) | — | 73.5% |
Multimodal Gemma 4 31B IT leads
Gemma 4 31B IT: 41.6 (#34), o4-mini: 40.2 (#49)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| LMArena Vision | 1277 | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), o4-mini: 47.0 (#154)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| LMArena Non-English | 1431 | 1337 |
| LMArena Chinese | 1476 | 1354 |
| LMArena French | 1435 | 1364 |
| LMArena Russian | 1460 | 1334 |
| LMArena Spanish | 1444 | 1347 |
| LMArena German | — | 1336 |
| LMArena Japanese | — | 1308 |
| LMArena Korean | — | 1312 |
Instruction Following Too close to call
Gemma 4 31B IT: 75.5 (#61), o4-mini: 75.2 (#68)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1433 | 1321 |
| IFEval | — | 92.8% |
Long Context o4-mini leads
Gemma 4 31B IT: 44.2 (#71), o4-mini: 45.5 (#33)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| LMArena Longer Query | 1446 | 1315 |
| Fiction.LiveBench | — | 77.8% |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), o4-mini: 54.0 (#152)
| Benchmark | Gemma 4 31B IT | o4-mini |
|---|---|---|
| LMArena Text | 1443 | 1353 |
| LMArena Creative Writing | 1415 | 1294 |
| LMArena Multi-Turn | 1452 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 1368 | — |
| WildBench | — | 85.4% |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than o4-mini?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 41.6 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or o4-mini?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; o4-mini lists at $1.10 and $4.40.
Is Gemma 4 31B IT or o4-mini better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 40.9 in the Noometry coding category.
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 o4-mini share?
27 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and o4-mini has 60.