Model comparison
Gemma 4 31B IT vs o1
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 40.9 on the Noometry Index.
Last verified . 23 shared benchmarks.
GoogleSummary
- They share 23 benchmarks with published results for both. Gemma 4 31B IT scores higher in 5 categories and o1 in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where Gemma 4 31B IT leads 41.6 to 34.2.
- The biggest single-benchmark swing is SimpleQA Verified: 10.4% for Gemma 4 31B IT and 41.1% for o1.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $15 / $60 for o1.
- 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 | o1 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 40.9 |
| Released | 2026-04-02 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 33K | 100K |
| Input $ / M tokens | $0.09 | $15 |
| Output $ / M tokens | $0.34 | $60 |
| Results tracked | 35 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
Gemma 4 31B IT: 42.3 (#108), o1: 46.1 (#70)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| WeirdML | 52.3% | 47.6% |
| LMArena Coding | 1459 | 1367 |
| Aider Polyglot | — | 61.7% |
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| ALE-Bench | 925.5 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, o1: 24.6 (#117)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |
Reasoning Too close to call
Gemma 4 31B IT: 27.2 (#122), o1: 27.9 (#111)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| Chess Puzzles | 5% | 15% |
| LMArena Hard Prompts | 1448 | 1371 |
| DTBench | 82.7% | 74.7% |
| LMCA | 39.3% | 22.3% |
| Epoch Capabilities Index | 142.74 | 141.91 |
| SimpleBench | — | 41.7% |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| ARC-AGI-1 | — | 30.7% |
| CritPt | 1.4% | — |
| EnigmaEval | — | 5.7% |
| Thematic Generalization | 53% | — |
| LiveBench Reasoning | — | 91.6% |
| LiveBench Data Analysis | — | 65.5% |
| Surface Evolver Bench | 30.6% | — |
| LiveBench | — | 75.7% |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), o1: 36.1 (#175)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 73.3% |
| LMArena Math | 1465 | 1388 |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath (Feb 2025 set) | — | 9.3% |
Knowledge o1 leads
Gemma 4 31B IT: 37.9 (#151), o1: 41.5 (#110)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| GPQA Diamond | 75.8% | 76.8% |
| SimpleQA Verified | 10.4% | 41.1% |
| LMArena Expert | 1465 | 1361 |
| Humanity's Last Exam | — | 8% |
| Confabulations | — | 11.7% |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Gemma 4 31B IT leads
Gemma 4 31B IT: 41.6 (#34), o1: 34.2 (#93)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| LMArena Vision | 1277 | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| LMArena Document | 1425 | — |
| SpatialViz-Bench | — | 41.4% |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), o1: 48.6 (#142)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| LMArena Non-English | 1431 | 1358 |
| LMArena Chinese | 1476 | 1394 |
| LMArena French | 1435 | 1344 |
| LMArena Russian | 1460 | 1356 |
| LMArena Spanish | 1444 | 1345 |
| LMArena German | — | 1337 |
| LMArena Japanese | — | 1346 |
| LMArena Korean | — | 1396 |
Instruction Following Too close to call
Gemma 4 31B IT: 75.5 (#61), o1: 74.8 (#86)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
Long Context o1 leads
Gemma 4 31B IT: 44.2 (#71), o1: 50.3 (#9)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| LMArena Longer Query | 1446 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), o1: 55.6 (#144)
| Benchmark | Gemma 4 31B IT | o1 |
|---|---|---|
| LMArena Text | 1443 | 1366 |
| LMArena Creative Writing | 1415 | 1348 |
| LMArena Multi-Turn | 1452 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is Gemma 4 31B IT better than o1?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 40.9 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or o1?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; o1 lists at $15 and $60.
Is Gemma 4 31B IT or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 42.3 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 o1 share?
23 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and o1 has 52.