Model comparison
Gemini 3.8 Flash vs GPT-5.6 Sol
GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 61.8 on the Noometry Index. Gemini 3.8 Flash costs 5.3× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.
Last verified . 48 shared benchmarks.
GoogleSummary
- They share 48 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 5 categories and GPT-5.6 Sol in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Sol leads 85.6 to 65.3.
- The biggest single-benchmark swing is FrontierMath Tier 4: 22% for Gemini 3.8 Flash and 82.9% for GPT-5.6 Sol.
- Gemini 3.8 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $4 / $20 for GPT-5.6 Sol.
- GPT-5.6 Sol accepts more context: 1.05M tokens versus 1.05M.
Side by side
| Gemini 3.8 Flash | GPT-5.6 Sol | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 61.8 | 65.0 |
| Released | 2026-09-02 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.75 | $4 |
| Output $ / M tokens | $3.75 | $20 |
| Results tracked | 50 | 65 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Sol leads
Gemini 3.8 Flash: 59.2 (#15), GPT-5.6 Sol: 65.1 (#7)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| DeepSWE | 73.8% | 72.7% |
| FrontierCode | 41.2% | 47.5% |
| CursorBench | 39.6% | 41.7% |
| LMArena WebDev | 1584 | 1618 |
| FrontierSWE | 19.6% | 32.2% |
| SciCode | 56.6% | 57.1% |
| WeirdML | 84.8% | 89.4% |
| LMArena Coding | 1510 | 1498 |
| ALE-Bench | 1,270 | 2,177 |
| GSO | — | 76.5% |
| MirrorCode | — | 20% |
Agentic & Tool Use GPT-5.6 Sol leads
Gemini 3.8 Flash: 41.8 (#21), GPT-5.6 Sol: 50.3 (#7)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| APEX-Agents | 64.3% | 51.4% |
| GDP.pdf | 23.4% | 30.7% |
| Vending-Bench 2 | 5,094 | 9,619 |
| OSWorld 2.0 | — | 27.3% |
| Remote Labor Index | 5.8% | — |
| τ²-bench Banking | — | 46.9% |
| PostTrainBench | — | 36.2% |
| BALROG | — | 60% |
| GBAEval | — | 52.6% |
| LMArena Search | — | 1257 |
Reasoning Gemini 3.8 Flash leads
Gemini 3.8 Flash: 76.9 (#5), GPT-5.6 Sol: 74.8 (#8)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| ARC-AGI-2 | 89.2% | 92.5% |
| NYT Connections (extended) | 97.4% | 93.8% |
| ARC-AGI-1 | 98.5% | 97.5% |
| CritPt | 18.3% | 32.3% |
| Chess Puzzles | 61% | 64% |
| LMArena Hard Prompts | 1508 | 1484 |
| Mystery Game Puzzles | 47% | 58% |
| DTBench | 95.7% | 96% |
| LMCA | 52.9% | 59.2% |
| Surface Evolver Bench | 76.9% | 93.1% |
| Epoch Capabilities Index | 156.71 | 161.66 |
| SimpleBench | — | 71.7% |
| Kagi LLM Benchmark | — | 67% |
| EnigmaEval | — | 37.1% |
| EBR-Bench | — | 44.8% |
| Bench to the Future 3 | — | 0.14 |
Math GPT-5.6 Sol leads
Gemini 3.8 Flash: 65.3 (#28), GPT-5.6 Sol: 85.6 (#9)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.4% | 89.1% |
| FrontierMath Tier 4 | 22% | 82.9% |
| OTIS Mock AIME 2024-2025 | 98.9% | 100% |
| ProofBench | 48% | 83% |
| LMArena Math | 1528 | 1474 |
| FrontierMath Erdős | — | 0% |
Knowledge Gemini 3.8 Flash leads
Gemini 3.8 Flash: 74.8 (#2), GPT-5.6 Sol: 64.3 (#18)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| GPQA Diamond | 95.4% | 93.5% |
| SimpleQA Verified | 69.7% | 69.7% |
| LMArena Expert | 1524 | 1516 |
| Humanity's Last Exam | 44.5% | — |
| Vectara Hallucination Rate | — | 12.4% |
Multimodal GPT-5.6 Sol leads
Gemini 3.8 Flash: 40.7 (#45), GPT-5.6 Sol: 48.6 (#9)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| LMArena Vision | 1314 | 1281 |
| Blueprint-Bench 2 | 38.6% | 33.6% |
| Furniture Assembly | 31.7% | 56.7% |
| LMArena Document | — | 1483 |
Multilingual Gemini 3.8 Flash leads
Gemini 3.8 Flash: 58.0 (#5), GPT-5.6 Sol: 55.3 (#32)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| LMArena Non-English | 1491 | 1452 |
| LMArena Chinese | 1554 | 1527 |
| LMArena French | 1498 | 1477 |
| LMArena German | 1493 | 1476 |
| LMArena Japanese | 1502 | 1471 |
| LMArena Korean | 1459 | 1442 |
| LMArena Russian | 1515 | 1468 |
| LMArena Spanish | 1485 | 1441 |
Instruction Following Too close to call
Gemini 3.8 Flash: 78.0 (#13), GPT-5.6 Sol: 77.7 (#16)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| LMArena Instruction Following | 1490 | 1482 |
Long Context Too close to call
Gemini 3.8 Flash: 46.3 (#24), GPT-5.6 Sol: 45.4 (#42)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| LMArena Longer Query | 1508 | 1480 |
Writing & Preference GPT-5.6 Sol leads
Gemini 3.8 Flash: 72.2 (#15), GPT-5.6 Sol: 73.3 (#12)
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| LMArena Text | 1499 | 1457 |
| LMArena Creative Writing | 1492 | 1448 |
| EQ-Bench Creative Writing | 1748 | 1972 |
| LMArena Multi-Turn | 1501 | 1460 |
| EQ-Bench 4 | — | 1250 |
Frequently asked questions
Is Gemini 3.8 Flash better than GPT-5.6 Sol?
GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 61.8 on the Noometry Index. Gemini 3.8 Flash costs 5.3× 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, Gemini 3.8 Flash or GPT-5.6 Sol?
Gemini 3.8 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; GPT-5.6 Sol lists at $4 and $20.
Is Gemini 3.8 Flash or GPT-5.6 Sol better for coding?
GPT-5.6 Sol scores higher on coding benchmarks: 65.1 versus 59.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Sol does, with 1.05M tokens against 1.05M.
How many benchmarks do Gemini 3.8 Flash and GPT-5.6 Sol share?
48 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and GPT-5.6 Sol has 65.