Model comparison
Gemini 3.1 Pro Preview vs GPT-5.4 nano
Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 41.9 on the Noometry Index. GPT-5.4 nano costs 9.7× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.
Last verified . 39 shared benchmarks.
GoogleSummary
- They share 39 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 8 categories and GPT-5.4 nano in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 23.7.
- The biggest single-benchmark swing is ARC-AGI-2: 77.1% for Gemini 3.1 Pro Preview and 5.7% for GPT-5.4 nano.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
- Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 400K.
Side by side
| Gemini 3.1 Pro Preview | GPT-5.4 nano | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 56.7 | 41.9 |
| Released | 2026-02-19 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 400K |
| Max output | 66K | 128K |
| Input $ / M tokens | $2 | $0.20 |
| Output $ / M tokens | $12 | $1.25 |
| Results tracked | 71 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
Gemini 3.1 Pro Preview: 42.5 (#99), GPT-5.4 nano: 43.6 (#84)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| SciCode | 58.9% | 46.9% |
| WeirdML | 72.1% | 49.2% |
| LMArena Coding | 1484 | 1405 |
| ALE-Bench | 1,161 | 1,005 |
| SWE-bench Verified | 75.6% | — |
| DeepSWE | 11.7% | — |
| LMArena WebDev | 1447 | — |
| GSO | 22.6% | — |
| MirrorCode | 8.9% | — |
| AlgoTune | 2.02 | — |
Agentic & Tool Use Not comparable
Gemini 3.1 Pro Preview: 37.7 (#34), GPT-5.4 nano: —
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| Terminal-Bench | 80.2% | — |
| APEX-Agents | 35.3% | — |
| τ²-bench Banking | 26% | — |
| DeepResearch Bench | 47.8% | — |
| PostTrainBench | 22% | — |
| BALROG | 57% | — |
| ExploitBench | 26.1% | — |
| GBAEval | 0.8% | — |
| GDP.pdf | 17% | — |
| LMArena Search | 1211 | — |
| METR Time Horizons | 77% | — |
| Vending-Bench 2 | 3,774 | — |
Reasoning Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 71.7 (#12), GPT-5.4 nano: 23.7 (#173)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| ARC-AGI-2 | 77.1% | 5.7% |
| ARC-AGI-1 | 98% | 51.5% |
| CritPt | 17.7% | 9.3% |
| Chess Puzzles | 55% | 30% |
| LMArena Hard Prompts | 1485 | 1381 |
| Mystery Game Puzzles | 34% | 9% |
| DTBench | 97.1% | 80.3% |
| LMCA | 53.8% | 36.9% |
| Epoch Capabilities Index | 154.77 | 145.81 |
| ForecastBench | 59 | 57.3 |
| SimpleBench | 79.6% | — |
| Kagi LLM Benchmark | — | 39.7% |
| NYT Connections (extended) | 97.4% | — |
| EnigmaEval | 36.8% | — |
| Thematic Generalization | 79.4% | — |
| EBR-Bench | 14.3% | — |
Math Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 62.1 (#34), GPT-5.4 nano: 40.9 (#88)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.6% | 44.9% |
| FrontierMath Tier 4 | 26.8% | 12.2% |
| OTIS Mock AIME 2024-2025 | 95.6% | 87.8% |
| ProofBench | 26% | 5% |
| LMArena Math | 1485 | 1406 |
| FrontierMath (Feb 2025 set) | 36.9% | 25.9% |
| FrontierMath Tier 4 (v1) | 16.7% | 6.3% |
| MathArena Final-Answer Competitions | 86.5% | — |
Knowledge Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 71.8 (#3), GPT-5.4 nano: 41.9 (#103)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| GPQA Diamond | 94.4% | 78.5% |
| SimpleQA Verified | 73.5% | 11.7% |
| Vectara Hallucination Rate | 10.4% | 3.1% |
| LMArena Expert | 1485 | 1396 |
| Humanity's Last Exam | 46.4% | — |
Multimodal Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 37.9 (#69), GPT-5.4 nano: 36.7 (#78)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | 1296 | 1196 |
| Blueprint-Bench 2 | 26.5% | — |
| Furniture Assembly | 26.7% | — |
| LMArena Document | 1444 | — |
Multilingual Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 57.0 (#12), GPT-5.4 nano: 48.6 (#140)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1477 | 1359 |
| LMArena Chinese | 1529 | 1392 |
| LMArena French | 1487 | 1396 |
| LMArena German | 1491 | 1367 |
| LMArena Japanese | 1493 | 1343 |
| LMArena Korean | 1455 | 1320 |
| LMArena Russian | 1498 | 1363 |
| LMArena Spanish | 1479 | 1371 |
Instruction Following Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 77.0 (#32), GPT-5.4 nano: 71.9 (#144)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1466 | 1362 |
Long Context Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 47.4 (#18), GPT-5.4 nano: 41.6 (#137)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1483 | 1366 |
| CL-bench | 20.8% | — |
| CL-bench Life | 16.9% | — |
Writing & Preference Gemini 3.1 Pro Preview leads
Gemini 3.1 Pro Preview: 66.1 (#37), GPT-5.4 nano: 55.7 (#142)
| Benchmark | Gemini 3.1 Pro Preview | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1481 | 1372 |
| LMArena Creative Writing | 1482 | 1314 |
| LMArena Multi-Turn | 1488 | 1382 |
| EQ-Bench Creative Writing | 1491 | — |
| EQ-Bench 4 | 1142 | — |
Frequently asked questions
Is Gemini 3.1 Pro Preview better than GPT-5.4 nano?
Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 41.9 on the Noometry Index. GPT-5.4 nano costs 9.7× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.1 Pro Preview or GPT-5.4 nano?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.
Is Gemini 3.1 Pro Preview or GPT-5.4 nano better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 42.5 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.1 Pro Preview does, with 1.05M tokens against 400K.
How many benchmarks do Gemini 3.1 Pro Preview and GPT-5.4 nano share?
39 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and GPT-5.4 nano has 40.