Model comparison
Gemma 4 31B IT vs GPT-5.6 Sol
GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 52× less per token, which makes it the better buy when GPT-5.6 Sol'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-5.6 Sol in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.6 Sol leads 74.8 to 27.2.
- The biggest single-benchmark swing is Surface Evolver Bench: 30.6% for Gemma 4 31B IT and 93.1% for GPT-5.6 Sol.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $4 / $20 for GPT-5.6 Sol.
- GPT-5.6 Sol 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-5.6 Sol | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 65.0 |
| Released | 2026-04-02 | 2026-07-09 |
| Weights | Open | Proprietary |
| Context window | 262K | 1.05M |
| Max output | 33K | 128K |
| Input $ / M tokens | $0.09 | $4 |
| Output $ / M tokens | $0.34 | $20 |
| Results tracked | 35 | 65 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Sol leads
Gemma 4 31B IT: 42.3 (#108), GPT-5.6 Sol: 65.1 (#7)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| LMArena WebDev | 1366 | 1618 |
| SciCode | 43.4% | 57.1% |
| WeirdML | 52.3% | 89.4% |
| LMArena Coding | 1459 | 1498 |
| ALE-Bench | 925.5 | 2,177 |
| DeepSWE | — | 72.7% |
| FrontierCode | — | 47.5% |
| CursorBench | — | 41.7% |
| FrontierSWE | — | 32.2% |
| GSO | — | 76.5% |
| MirrorCode | — | 20% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, GPT-5.6 Sol: 50.3 (#7)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| APEX-Agents | — | 51.4% |
| OSWorld 2.0 | — | 27.3% |
| τ²-bench Banking | — | 46.9% |
| PostTrainBench | — | 36.2% |
| BALROG | — | 60% |
| GBAEval | — | 52.6% |
| GDP.pdf | — | 30.7% |
| LMArena Search | — | 1257 |
| Vending-Bench 2 | — | 9,619 |
Reasoning GPT-5.6 Sol leads
Gemma 4 31B IT: 27.2 (#122), GPT-5.6 Sol: 74.8 (#8)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 67% |
| NYT Connections (extended) | 70.6% | 93.8% |
| CritPt | 1.4% | 32.3% |
| Chess Puzzles | 5% | 64% |
| LMArena Hard Prompts | 1448 | 1484 |
| DTBench | 82.7% | 96% |
| LMCA | 39.3% | 59.2% |
| Surface Evolver Bench | 30.6% | 93.1% |
| Epoch Capabilities Index | 142.74 | 161.66 |
| ARC-AGI-2 | — | 92.5% |
| SimpleBench | — | 71.7% |
| ARC-AGI-1 | — | 97.5% |
| EnigmaEval | — | 37.1% |
| Thematic Generalization | 53% | — |
| EBR-Bench | — | 44.8% |
| Mystery Game Puzzles | — | 58% |
| Bench to the Future 3 | — | 0.14 |
Math GPT-5.6 Sol leads
Gemma 4 31B IT: 43.2 (#81), GPT-5.6 Sol: 85.6 (#9)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 100% |
| LMArena Math | 1465 | 1474 |
| FrontierMath (Tiers 1-3) | — | 89.1% |
| FrontierMath Tier 4 | — | 82.9% |
| ProofBench | — | 83% |
| FrontierMath Erdős | — | 0% |
Knowledge GPT-5.6 Sol leads
Gemma 4 31B IT: 37.9 (#151), GPT-5.6 Sol: 64.3 (#18)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| GPQA Diamond | 75.8% | 93.5% |
| SimpleQA Verified | 10.4% | 69.7% |
| Vectara Hallucination Rate | 7.4% | 12.4% |
| LMArena Expert | 1465 | 1516 |
Multimodal GPT-5.6 Sol leads
Gemma 4 31B IT: 41.6 (#34), GPT-5.6 Sol: 48.6 (#9)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| LMArena Vision | 1277 | 1281 |
| LMArena Document | 1425 | 1483 |
| Blueprint-Bench 2 | — | 33.6% |
| Furniture Assembly | — | 56.7% |
Multilingual GPT-5.6 Sol leads
Gemma 4 31B IT: 53.8 (#57), GPT-5.6 Sol: 55.3 (#32)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| LMArena Non-English | 1431 | 1452 |
| LMArena Chinese | 1476 | 1527 |
| LMArena French | 1435 | 1477 |
| LMArena Russian | 1460 | 1468 |
| LMArena Spanish | 1444 | 1441 |
| LMArena German | — | 1476 |
| LMArena Japanese | — | 1471 |
| LMArena Korean | — | 1442 |
Instruction Following GPT-5.6 Sol leads
Gemma 4 31B IT: 75.5 (#61), GPT-5.6 Sol: 77.7 (#16)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| LMArena Instruction Following | 1433 | 1482 |
Long Context GPT-5.6 Sol leads
Gemma 4 31B IT: 44.2 (#71), GPT-5.6 Sol: 45.4 (#42)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| LMArena Longer Query | 1446 | 1480 |
Writing & Preference GPT-5.6 Sol leads
Gemma 4 31B IT: 60.5 (#96), GPT-5.6 Sol: 73.3 (#12)
| Benchmark | Gemma 4 31B IT | GPT-5.6 Sol |
|---|---|---|
| LMArena Text | 1443 | 1457 |
| LMArena Creative Writing | 1415 | 1448 |
| EQ-Bench Creative Writing | 1368 | 1972 |
| EQ-Bench 4 | 1120 | 1250 |
| LMArena Multi-Turn | 1452 | 1460 |
Frequently asked questions
Is Gemma 4 31B IT better than GPT-5.6 Sol?
GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 52× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 31B IT or GPT-5.6 Sol?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; GPT-5.6 Sol lists at $4 and $20.
Is Gemma 4 31B IT or GPT-5.6 Sol better for coding?
GPT-5.6 Sol scores higher on coding benchmarks: 65.1 versus 42.3 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Sol does, with 1.05M tokens against 262K.
How many benchmarks do Gemma 4 31B IT and GPT-5.6 Sol share?
34 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and GPT-5.6 Sol has 65.