Model comparison
Gemma 4 31B IT vs GPT-4 Turbo
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 30.5 on the Noometry Index.
Last verified . 22 shared benchmarks.
GoogleSummary
- They share 22 benchmarks with published results for both. Gemma 4 31B IT scores higher in 9 categories and GPT-4 Turbo in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 43.2 to 9.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.3% for Gemma 4 31B IT and 6.7% for GPT-4 Turbo.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- Gemma 4 31B IT accepts more context: 262K tokens versus 128K.
- Gemma 4 31B IT has downloadable open weights; the other is API-only.
Side by side
| Gemma 4 31B IT | GPT-4 Turbo | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 30.5 |
| Released | 2026-04-02 | 2023-11-06 |
| Weights | Open | Proprietary |
| Context window | 262K | 128K |
| Max output | 33K | 4K |
| Input $ / M tokens | $0.09 | $10 |
| Output $ / M tokens | $0.34 | $30 |
| Results tracked | 35 | 36 |
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 Turbo: 33.8 (#249)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| WeirdML | 52.3% | 18% |
| LMArena Coding | 1459 | 1268 |
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| BigCodeBench Instruct | — | 48.2% |
| BigCodeBench Complete | — | 58.2% |
| ALE-Bench | 925.5 | — |
| HumanEval+ | — | 86.6% |
| MBPP+ | — | 73.3% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, GPT-4 Turbo: —
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| METR Time Horizons | — | 36.7% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), GPT-4 Turbo: 15.3 (#317)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| Chess Puzzles | 5% | 6% |
| LMArena Hard Prompts | 1448 | 1251 |
| DTBench | 82.7% | 61.6% |
| LMCA | 39.3% | 9.8% |
| Epoch Capabilities Index | 142.74 | 127.25 |
| SimpleBench | — | 25.1% |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| CritPt | 1.4% | — |
| Thematic Generalization | 53% | — |
| Surface Evolver Bench | 30.6% | — |
| ForecastBench | — | 59.4 |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), GPT-4 Turbo: 9.0 (#322)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 6.7% |
| LMArena Math | 1465 | 1272 |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| MATH Level 5 | — | 46.7% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#151), GPT-4 Turbo: 24.3 (#268)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| GPQA Diamond | 75.8% | 46.6% |
| LMArena Expert | 1465 | 1223 |
| SimpleQA Verified | 10.4% | — |
| Confabulations | — | 28.4% |
| Vectara Hallucination Rate | 7.4% | — |
| MMLU | — | 81.3% |
Multimodal Gemma 4 31B IT leads
Gemma 4 31B IT: 41.6 (#34), GPT-4 Turbo: 30.6 (#110)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| LMArena Vision | 1277 | 1090 |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), GPT-4 Turbo: 40.5 (#216)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| LMArena Non-English | 1431 | 1245 |
| LMArena Chinese | 1476 | 1242 |
| LMArena French | 1435 | 1276 |
| LMArena Russian | 1460 | 1259 |
| LMArena Spanish | 1444 | 1260 |
| LMArena German | — | 1259 |
| LMArena Japanese | — | 1194 |
| LMArena Korean | — | 1187 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), GPT-4 Turbo: 65.8 (#216)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| LMArena Instruction Following | 1433 | 1249 |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), GPT-4 Turbo: 38.0 (#206)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| LMArena Longer Query | 1446 | 1254 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), GPT-4 Turbo: 47.7 (#206)
| Benchmark | Gemma 4 31B IT | GPT-4 Turbo |
|---|---|---|
| LMArena Text | 1443 | 1272 |
| LMArena Creative Writing | 1415 | 1269 |
| LMArena Multi-Turn | 1452 | 1267 |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than GPT-4 Turbo?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 30.5 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or GPT-4 Turbo?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is Gemma 4 31B IT or GPT-4 Turbo better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 128K.
How many benchmarks do Gemma 4 31B IT and GPT-4 Turbo share?
22 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and GPT-4 Turbo has 36.