Leaderboard / Model
Gemini 3.6 Flash
Google DeepMind · Closed weights · Released Jul 21, 2026 · $0.75 in · $3.75 out per 1M tokens
50Overall · rank 19
Score by job
Coding
47 · rank 22 · 3 of 5 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| DeepSWE | 46.7% | gemini-3.6-flash_high | -0.67 |
| FrontierCode | 34.4% | gemini-3.6-flash_unknown | -0.54 |
| LMArena Coding | 1491 rating | gemini-3.6-flash-high | 0.01 |
Agents
Not enough data · 1 of 4 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| APEX-Agents | 46.9% | gemini-3.6-flash_unknown | -0.52 |
Reasoning
62 · rank 13 · 2 of 4 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| GPQA Diamond | 94.1% | gemini-3.6-flash_high | 1.03 |
| ARC-AGI-2 | 60.4% | gemini-3.6-flash_high | -0.16 |
Math
37 · rank 23 · 5 of 5 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| FrontierMath Tiers 1–3 | 58.9% | gemini-3.6-flash_high | -0.62 |
| FrontierMath Tier 4 | 22.0% | gemini-3.6-flash_high | -0.80 |
| OTIS Mock AIME | 94.2% | gemini-3.6-flash_high | -0.34 |
| ProofBench | 36.0% | gemini-3.6-flash_unknown | -0.46 |
| LMArena Math | 1506 rating | gemini-3.6-flash-high | 0.97 |
Writing
67 · rank 16 · 2 of 2 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| LMArena Creative Writing | 1464 rating | gemini-3.6-flash-high | 0.39 |
| LMArena Instruction Following | 1467 rating | gemini-3.6-flash-high | 0.09 |
Design
37 · rank 25 · 1 of 1 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| LMArena WebDev | 1537 rating | gemini-3.6-flash-high | -0.21 |