Model comparison
Gemini 3.6 Flash vs GPT-5.5
GPT-5.5 is the stronger model overall, scoring 63.4 to 54.1 on the Noometry Index. Gemini 3.6 Flash costs 7.5× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Last verified . 46 shared benchmarks.
GoogleSummary
- They share 46 benchmarks with published results for both. Gemini 3.6 Flash scores higher in 2 categories and GPT-5.5 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.5 leads 81.7 to 57.3.
- The biggest single-benchmark swing is FrontierMath Tier 4: 22% for Gemini 3.6 Flash and 72.5% for GPT-5.5.
- Gemini 3.6 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 1.05M.
Side by side
| Gemini 3.6 Flash | GPT-5.5 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 54.1 | 63.4 |
| Released | 2026-07-21 | 2026-04-23 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.75 | $5 |
| Output $ / M tokens | $3.75 | $30 |
| Results tracked | 46 | 71 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 leads
Gemini 3.6 Flash: 50.0 (#48), GPT-5.5: 58.2 (#17)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| DeepSWE | 46.7% | 67% |
| FrontierCode | 34.4% | 43% |
| LMArena WebDev | 1538 | 1513 |
| SciCode | 52.7% | 56.1% |
| WeirdML | 56.1% | 84.9% |
| LMArena Coding | 1491 | 1494 |
| ALE-Bench | 715.52 | 1,943 |
| SWE-bench Verified | — | 80.6% |
| GSO | — | 40.2% |
| MirrorCode | — | 10% |
Agentic & Tool Use GPT-5.5 leads
Gemini 3.6 Flash: 32.3 (#65), GPT-5.5: 50.7 (#6)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| APEX-Agents | 46.9% | 55.1% |
| GDP.pdf | 14% | 26% |
| Terminal-Bench | — | 84.7% |
| 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% |
| LMArena Search | — | 1242 |
| Vending-Bench 2 | — | 7,524 |
Reasoning GPT-5.5 leads
Gemini 3.6 Flash: 58.8 (#22), GPT-5.5: 72.8 (#11)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| ARC-AGI-2 | 60.4% | 85% |
| NYT Connections (extended) | 89% | 96.2% |
| ARC-AGI-1 | 91.2% | 95% |
| CritPt | 10.6% | 27.1% |
| Chess Puzzles | 43% | 54% |
| LMArena Hard Prompts | 1485 | 1489 |
| Mystery Game Puzzles | 30% | 56% |
| DTBench | 95.5% | 96% |
| LMCA | 44.9% | 54.3% |
| Epoch Capabilities Index | 154.25 | 159.1 |
| SimpleBench | — | 69% |
| Kagi LLM Benchmark | — | 88.8% |
| EBR-Bench | — | 34.3% |
| Surface Evolver Bench | — | 88.1% |
| Bench to the Future 3 | — | 0.14 |
| ForecastBench | — | 60.6 |
Math GPT-5.5 leads
Gemini 3.6 Flash: 57.3 (#40), GPT-5.5: 81.7 (#11)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 58.9% | 85.3% |
| FrontierMath Tier 4 | 22% | 72.5% |
| MathArena Final-Answer Competitions | 70.8% | 94.3% |
| OTIS Mock AIME 2024-2025 | 94.2% | 100% |
| ProofBench | 36% | 50% |
| LMArena Math | 1505 | 1486 |
| FrontierMath (Feb 2025 set) | — | 51.7% |
| FrontierMath Erdős | — | 0% |
| FrontierMath Tier 4 (v1) | — | 35.4% |
Knowledge Gemini 3.6 Flash leads
Gemini 3.6 Flash: 67.8 (#8), GPT-5.5: 64.4 (#17)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 94.1% | 94% |
| SimpleQA Verified | 66.2% | 63% |
| LMArena Expert | 1488 | 1508 |
| Vectara Hallucination Rate | — | 9.3% |
Multimodal GPT-5.5 leads
Gemini 3.6 Flash: 38.5 (#64), GPT-5.5: 46.9 (#12)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| LMArena Vision | 1298 | 1297 |
| Blueprint-Bench 2 | 31.2% | 36.2% |
| Furniture Assembly | 23.3% | 44.2% |
| LMArena Document | 1456 | 1486 |
Multilingual Too close to call
Gemini 3.6 Flash: 56.5 (#19), GPT-5.5: 56.4 (#20)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1469 | 1467 |
| LMArena Chinese | 1531 | 1533 |
| LMArena French | 1504 | 1486 |
| LMArena German | 1478 | 1480 |
| LMArena Japanese | 1476 | 1498 |
| LMArena Korean | 1431 | 1460 |
| LMArena Russian | 1487 | 1473 |
| LMArena Spanish | 1475 | 1468 |
Instruction Following Too close to call
Gemini 3.6 Flash: 77.0 (#33), GPT-5.5: 77.5 (#18)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1466 | 1479 |
Long Context GPT-5.5 leads
Gemini 3.6 Flash: 45.1 (#50), GPT-5.5: 48.3 (#12)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| LMArena Longer Query | 1474 | 1484 |
| CL-bench Life | — | 22.2% |
Writing & Preference GPT-5.5 leads
Gemini 3.6 Flash: 68.2 (#27), GPT-5.5: 72.7 (#13)
| Benchmark | Gemini 3.6 Flash | GPT-5.5 |
|---|---|---|
| LMArena Text | 1479 | 1472 |
| LMArena Creative Writing | 1465 | 1455 |
| EQ-Bench Creative Writing | 1604 | 1844 |
| LMArena Multi-Turn | 1481 | 1476 |
| EQ-Bench 4 | — | 1315 |
Frequently asked questions
Is Gemini 3.6 Flash better than GPT-5.5?
GPT-5.5 is the stronger model overall, scoring 63.4 to 54.1 on the Noometry Index. Gemini 3.6 Flash costs 7.5× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.6 Flash or GPT-5.5?
Gemini 3.6 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; GPT-5.5 lists at $5 and $30.
Is Gemini 3.6 Flash or GPT-5.5 better for coding?
GPT-5.5 scores higher on coding benchmarks: 58.2 versus 50.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 1.05M.
How many benchmarks do Gemini 3.6 Flash and GPT-5.5 share?
46 benchmarks have published results for both models. Gemini 3.6 Flash has 46 scored results on Noometry and GPT-5.5 has 71.