Model comparison
Gemma 4 31B IT vs GPT-5.5
GPT-5.5 is the stronger model overall, scoring 63.4 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 74× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
GoogleSummary
- They share 34 benchmarks with published results for both. Gemma 4 31B IT scores higher in 0 categories and GPT-5.5 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.5 leads 72.8 to 27.2.
- The biggest single-benchmark swing is Surface Evolver Bench: 30.6% for Gemma 4 31B IT and 88.1% for GPT-5.5.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 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-5.5 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 63.4 |
| Released | 2026-04-02 | 2026-04-23 |
| Weights | Open | Proprietary |
| Context window | 262K | 1.05M |
| Max output | 33K | 128K |
| Input $ / M tokens | $0.09 | $5 |
| Output $ / M tokens | $0.34 | $30 |
| Results tracked | 35 | 71 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 leads
Gemma 4 31B IT: 42.3 (#108), GPT-5.5: 58.2 (#17)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| LMArena WebDev | 1366 | 1513 |
| SciCode | 43.4% | 56.1% |
| WeirdML | 52.3% | 84.9% |
| LMArena Coding | 1459 | 1494 |
| ALE-Bench | 925.5 | 1,943 |
| SWE-bench Verified | — | 80.6% |
| DeepSWE | — | 67% |
| FrontierCode | — | 43% |
| GSO | — | 40.2% |
| MirrorCode | — | 10% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, GPT-5.5: 50.7 (#6)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| Terminal-Bench | — | 84.7% |
| APEX-Agents | — | 55.1% |
| OSWorld 2.0 | — | 13% |
| Remote Labor Index | — | 6.3% |
| τ²-bench Banking | — | 44.6% |
| DeepResearch Bench | — | 54% |
| PostTrainBench | — | 27.2% |
| ExploitBench | — | 47.4% |
| GBAEval | — | 53.2% |
| GDP.pdf | — | 26% |
| LMArena Search | — | 1242 |
| Vending-Bench 2 | — | 7,524 |
Reasoning GPT-5.5 leads
Gemma 4 31B IT: 27.2 (#122), GPT-5.5: 72.8 (#11)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 88.8% |
| NYT Connections (extended) | 70.6% | 96.2% |
| CritPt | 1.4% | 27.1% |
| Chess Puzzles | 5% | 54% |
| LMArena Hard Prompts | 1448 | 1489 |
| DTBench | 82.7% | 96% |
| LMCA | 39.3% | 54.3% |
| Surface Evolver Bench | 30.6% | 88.1% |
| Epoch Capabilities Index | 142.74 | 159.1 |
| ARC-AGI-2 | — | 85% |
| SimpleBench | — | 69% |
| ARC-AGI-1 | — | 95% |
| Thematic Generalization | 53% | — |
| EBR-Bench | — | 34.3% |
| Mystery Game Puzzles | — | 56% |
| Bench to the Future 3 | — | 0.14 |
| ForecastBench | — | 60.6 |
Math GPT-5.5 leads
Gemma 4 31B IT: 43.2 (#81), GPT-5.5: 81.7 (#11)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 100% |
| LMArena Math | 1465 | 1486 |
| FrontierMath (Tiers 1-3) | — | 85.3% |
| FrontierMath Tier 4 | — | 72.5% |
| MathArena Final-Answer Competitions | — | 94.3% |
| ProofBench | — | 50% |
| FrontierMath (Feb 2025 set) | — | 51.7% |
| FrontierMath Erdős | — | 0% |
| FrontierMath Tier 4 (v1) | — | 35.4% |
Knowledge GPT-5.5 leads
Gemma 4 31B IT: 37.9 (#151), GPT-5.5: 64.4 (#17)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 75.8% | 94% |
| SimpleQA Verified | 10.4% | 63% |
| Vectara Hallucination Rate | 7.4% | 9.3% |
| LMArena Expert | 1465 | 1508 |
Multimodal GPT-5.5 leads
Gemma 4 31B IT: 41.6 (#34), GPT-5.5: 46.9 (#12)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| LMArena Vision | 1277 | 1297 |
| LMArena Document | 1425 | 1486 |
| Blueprint-Bench 2 | — | 36.2% |
| Furniture Assembly | — | 44.2% |
Multilingual GPT-5.5 leads
Gemma 4 31B IT: 53.8 (#57), GPT-5.5: 56.4 (#20)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1431 | 1467 |
| LMArena Chinese | 1476 | 1533 |
| LMArena French | 1435 | 1486 |
| LMArena Russian | 1460 | 1473 |
| LMArena Spanish | 1444 | 1468 |
| LMArena German | — | 1480 |
| LMArena Japanese | — | 1498 |
| LMArena Korean | — | 1460 |
Instruction Following GPT-5.5 leads
Gemma 4 31B IT: 75.5 (#61), GPT-5.5: 77.5 (#18)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1479 |
Long Context GPT-5.5 leads
Gemma 4 31B IT: 44.2 (#71), GPT-5.5: 48.3 (#12)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| LMArena Longer Query | 1446 | 1484 |
| CL-bench Life | — | 22.2% |
Writing & Preference GPT-5.5 leads
Gemma 4 31B IT: 60.5 (#96), GPT-5.5: 72.7 (#13)
| Benchmark | Gemma 4 31B IT | GPT-5.5 |
|---|---|---|
| LMArena Text | 1443 | 1472 |
| LMArena Creative Writing | 1415 | 1455 |
| EQ-Bench Creative Writing | 1368 | 1844 |
| EQ-Bench 4 | 1120 | 1315 |
| LMArena Multi-Turn | 1452 | 1476 |
Frequently asked questions
Is Gemma 4 31B IT better than GPT-5.5?
GPT-5.5 is the stronger model overall, scoring 63.4 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 74× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 31B IT or GPT-5.5?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; GPT-5.5 lists at $5 and $30.
Is Gemma 4 31B IT or GPT-5.5 better for coding?
GPT-5.5 scores higher on coding benchmarks: 58.2 versus 42.3 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 262K.
How many benchmarks do Gemma 4 31B IT and GPT-5.5 share?
34 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and GPT-5.5 has 71.