Model comparison
Gemma 4 31B IT vs o3
o3 is the stronger model overall, scoring 47.4 to 43.2 on the Noometry Index. Gemma 4 31B IT costs 23× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
GoogleSummary
- They share 27 benchmarks with published results for both. Gemma 4 31B IT scores higher in 3 categories and o3 in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o3 leads 54.5 to 37.9.
- The biggest single-benchmark swing is SimpleQA Verified: 10.4% for Gemma 4 31B IT and 49.4% for o3.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $2 / $8 for o3.
- 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 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.2 | 47.4 |
| Released | 2026-04-02 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 33K | 100K |
| Input $ / M tokens | $0.09 | $2 |
| Output $ / M tokens | $0.34 | $8 |
| Results tracked | 35 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
Gemma 4 31B IT: 41.9 (#98), o3: 47.3 (#57)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| WeirdML | 52.3% | 52.4% |
| LMArena Coding | 1459 | 1408 |
| ALE-Bench | 925.5 | 933.55 |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| LMArena WebDev | 1364 | — |
| SciCode | 43.4% | — |
| GSO | — | 8.8% |
| CadEval | — | 74% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, o3: 34.3 (#42)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
Gemma 4 31B IT: 26.4 (#119), o3: 32.1 (#66)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 67.6% |
| CritPt | 1.4% | 1.4% |
| Chess Puzzles | 5% | 38% |
| LMArena Hard Prompts | 1448 | 1402 |
| DTBench | 82.7% | 84.8% |
| LMCA | 39.3% | 39.7% |
| Epoch Capabilities Index | 142.74 | 146.88 |
| ARC-AGI-2 | — | 6.5% |
| SimpleBench | — | 53.1% |
| NYT Connections (extended) | 70.6% | — |
| ARC-AGI-1 | — | 60.8% |
| EnigmaEval | — | 13.1% |
| Thematic Generalization | 53% | — |
| Mystery Game Puzzles | — | 29% |
| Surface Evolver Bench | 30.6% | — |
| ForecastBench | — | 62.5 |
Math o3 leads
Gemma 4 31B IT: 41.3 (#72), o3: 48.9 (#49)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 84.4% |
| LMArena Math | 1464 | 1426 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
Gemma 4 31B IT: 37.9 (#141), o3: 54.5 (#47)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| GPQA Diamond | 75.8% | 81.8% |
| SimpleQA Verified | 10.4% | 49.4% |
| LMArena Expert | 1465 | 1403 |
| Humanity's Last Exam | — | 20.3% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| Vectara Hallucination Rate | 7.4% | — |
| GPQA (HELM) | — | 75.3% |
Multimodal Too close to call
Gemma 4 31B IT: 41.0 (#32), o3: 40.7 (#35)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| LMArena Vision | 1278 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 54.0 (#52), o3: 51.9 (#99)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| LMArena Non-English | 1431 | 1402 |
| LMArena Chinese | 1478 | 1438 |
| LMArena French | 1436 | 1430 |
| LMArena Russian | 1462 | 1407 |
| LMArena Spanish | 1443 | 1395 |
| LMArena German | — | 1420 |
| LMArena Japanese | — | 1404 |
| LMArena Korean | — | 1371 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.6 (#58), o3: 73.0 (#122)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1368 |
| IFEval | — | 86.9% |
Long Context o3 leads
Gemma 4 31B IT: 44.2 (#67), o3: 53.2 (#6)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| LMArena Longer Query | 1446 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
Gemma 4 31B IT: 60.7 (#92), o3: 63.7 (#62)
| Benchmark | Gemma 4 31B IT | o3 |
|---|---|---|
| LMArena Text | 1443 | 1410 |
| LMArena Creative Writing | 1416 | 1359 |
| EQ-Bench Creative Writing | 1368 | 1676 |
| LMArena Multi-Turn | 1452 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| WildBench | — | 86.1% |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than o3?
o3 is the stronger model overall, scoring 47.4 to 43.2 on the Noometry Index. Gemma 4 31B IT costs 23× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 31B IT or o3?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; o3 lists at $2 and $8.
Is Gemma 4 31B IT or o3 better for coding?
o3 scores higher on coding benchmarks: 47.3 versus 41.9 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 o3 share?
27 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and o3 has 63.