Model comparison
Gemma 4 31B IT vs GPT-4.1
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 35.9 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 9 categories and GPT-4.1 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 43.2 to 22.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.3% for Gemma 4 31B IT and 38.3% for GPT-4.1.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 262K.
- Gemma 4 31B IT has downloadable open weights; the other is API-only.
Side by side
| Gemma 4 31B IT | GPT-4.1 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 35.9 |
| Released | 2026-04-02 | 2025-04-14 |
| Weights | Open | Proprietary |
| Context window | 262K | 1.05M |
| Max output | 33K | 33K |
| Input $ / M tokens | $0.09 | $2 |
| Output $ / M tokens | $0.34 | $8 |
| Results tracked | 35 | 52 |
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), GPT-4.1: 34.4 (#238)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| WeirdML | 52.3% | 39% |
| LMArena Coding | 1459 | 1391 |
| ALE-Bench | 925.5 | 558.1 |
| SWE-bench Verified | — | 48.5% |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| CadEval | — | 42% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, GPT-4.1: 34.7 (#43)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 54% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), GPT-4.1: 11.7 (#339)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 52.3% |
| Chess Puzzles | 5% | 6% |
| LMArena Hard Prompts | 1448 | 1384 |
| DTBench | 82.7% | 68.3% |
| LMCA | 39.3% | 25.6% |
| Epoch Capabilities Index | 142.74 | 136.78 |
| ARC-AGI-2 | — | 0.4% |
| SimpleBench | — | 27% |
| NYT Connections (extended) | 70.6% | — |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 1.4% | — |
| EnigmaEval | — | 2.2% |
| Thematic Generalization | 53% | — |
| Surface Evolver Bench | 30.6% | — |
| ForecastBench | — | 61.5 |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), GPT-4.1: 22.3 (#280)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 38.3% |
| LMArena Math | 1465 | 1370 |
| FrontierMath (Tiers 1-3) | — | 6% |
| Omni-MATH | — | 47.1% |
| MATH Level 5 | — | 83% |
| FrontierMath (Feb 2025 set) | — | 5.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Too close to call
Gemma 4 31B IT: 37.9 (#151), GPT-4.1: 37.1 (#160)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| GPQA Diamond | 75.8% | 66.9% |
| SimpleQA Verified | 10.4% | 31.1% |
| Vectara Hallucination Rate | 7.4% | 5.6% |
| LMArena Expert | 1465 | 1364 |
| Humanity's Last Exam | — | 5.4% |
| MMLU-Pro | — | 81.1% |
| GPQA (HELM) | — | 65.9% |
Multimodal Gemma 4 31B IT leads
Gemma 4 31B IT: 41.6 (#34), GPT-4.1: 38.2 (#67)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| LMArena Vision | 1277 | 1211 |
| GeoBench | — | 72% |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), GPT-4.1: 49.4 (#133)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1431 | 1370 |
| LMArena Chinese | 1476 | 1382 |
| LMArena French | 1435 | 1382 |
| LMArena Russian | 1460 | 1377 |
| LMArena Spanish | 1444 | 1376 |
| LMArena German | — | 1381 |
| LMArena Japanese | — | 1319 |
| LMArena Korean | — | 1339 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), GPT-4.1: 71.3 (#153)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1367 |
| IFEval | — | 83.8% |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), GPT-4.1: 40.0 (#163)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1446 | 1385 |
| Fiction.LiveBench | — | 63.9% |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), GPT-4.1: 57.6 (#125)
| Benchmark | Gemma 4 31B IT | GPT-4.1 |
|---|---|---|
| LMArena Text | 1443 | 1383 |
| LMArena Creative Writing | 1415 | 1363 |
| EQ-Bench Creative Writing | 1368 | 1420 |
| LMArena Multi-Turn | 1452 | 1398 |
| WildBench | — | 85.4% |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than GPT-4.1?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 35.9 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or GPT-4.1?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; GPT-4.1 lists at $2 and $8.
Is Gemma 4 31B IT or GPT-4.1 better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 262K.
How many benchmarks do Gemma 4 31B IT and GPT-4.1 share?
27 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and GPT-4.1 has 52.