Model comparison
Gemma 4 31B IT vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 11× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
GoogleSummary
- They share 32 benchmarks with published results for both. Gemma 4 31B IT scores higher in 4 categories and GPT-5.4 mini in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5.4 mini leads 51.5 to 37.9.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 63.5% for Gemma 4 31B IT and 37.9% for GPT-5.4 mini.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- GPT-5.4 mini accepts more context: 400K tokens versus 262K.
- Gemma 4 31B IT has downloadable open weights; the other is API-only.
Side by side
| Gemma 4 31B IT | GPT-5.4 mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 45.0 |
| Released | 2026-04-02 | 2026-03-17 |
| Weights | Open | Proprietary |
| Context window | 262K | 400K |
| Max output | 33K | 128K |
| Input $ / M tokens | $0.09 | $0.75 |
| Output $ / M tokens | $0.34 | $4.50 |
| Results tracked | 35 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
Gemma 4 31B IT: 42.3 (#108), GPT-5.4 mini: 45.2 (#72)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| LMArena WebDev | 1366 | 1397 |
| SciCode | 43.4% | 49.9% |
| WeirdML | 52.3% | 60.3% |
| LMArena Coding | 1459 | 1438 |
| ALE-Bench | 925.5 | 1,189 |
| FrontierCode | — | 27% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, GPT-5.4 mini: 29.9 (#81)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| DeepResearch Bench | — | 36.3% |
Reasoning GPT-5.4 mini leads
Gemma 4 31B IT: 27.2 (#122), GPT-5.4 mini: 30.4 (#85)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 37.9% |
| NYT Connections (extended) | 70.6% | 61.8% |
| CritPt | 1.4% | 10% |
| Chess Puzzles | 5% | 24% |
| Thematic Generalization | 53% | 61.7% |
| LMArena Hard Prompts | 1448 | 1424 |
| DTBench | 82.7% | 80% |
| LMCA | 39.3% | 40.8% |
| Epoch Capabilities Index | 142.74 | 148.84 |
| ARC-AGI-2 | — | 18.9% |
| ARC-AGI-1 | — | 63.7% |
| Mystery Game Puzzles | — | 11% |
| Surface Evolver Bench | 30.6% | — |
| ForecastBench | — | 57 |
Math GPT-5.4 mini leads
Gemma 4 31B IT: 43.2 (#81), GPT-5.4 mini: 45.5 (#75)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 88.9% |
| LMArena Math | 1465 | 1419 |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 9.8% |
| ProofBench | — | 21% |
| FrontierMath (Feb 2025 set) | — | 28.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GPT-5.4 mini leads
Gemma 4 31B IT: 37.9 (#151), GPT-5.4 mini: 51.5 (#67)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 75.8% | 86.9% |
| SimpleQA Verified | 10.4% | 29.4% |
| Vectara Hallucination Rate | 7.4% | 5.5% |
| LMArena Expert | 1465 | 1435 |
Multimodal Gemma 4 31B IT leads
Gemma 4 31B IT: 41.6 (#34), GPT-5.4 mini: 39.7 (#56)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | 1277 | 1245 |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), GPT-5.4 mini: 51.9 (#96)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1431 | 1405 |
| LMArena Chinese | 1476 | 1446 |
| LMArena French | 1435 | 1440 |
| LMArena Russian | 1460 | 1417 |
| LMArena Spanish | 1444 | 1405 |
| LMArena German | — | 1409 |
| LMArena Japanese | — | 1374 |
| LMArena Korean | — | 1368 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), GPT-5.4 mini: 74.1 (#102)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1433 | 1405 |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), GPT-5.4 mini: 43.0 (#112)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1446 | 1407 |
Writing & Preference GPT-5.4 mini leads
Gemma 4 31B IT: 60.5 (#96), GPT-5.4 mini: 64.0 (#58)
| Benchmark | Gemma 4 31B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1443 | 1412 |
| LMArena Creative Writing | 1415 | 1370 |
| EQ-Bench Creative Writing | 1368 | 1665 |
| LMArena Multi-Turn | 1452 | 1429 |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 11× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 31B IT or GPT-5.4 mini?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.
Is Gemma 4 31B IT or GPT-5.4 mini better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 42.3 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 262K.
How many benchmarks do Gemma 4 31B IT and GPT-5.4 mini share?
32 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and GPT-5.4 mini has 46.