Gemini 3.5 Flash
ModelActiveby Google · family “gemini-flash”
Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution... (description from the OpenRouter listing)
in: textin: imagein: videoin: filein: audioout: textReasoningTool useStructured output
Epoch Capabilities Index
154.6 #30 of 268
90% CI 152.6 – 157.2
Released
19 May 2026
Context window
1.05M tokens
Max output
66K tokens
Input $ / 1M tokens
$1.50
Output $ / 1M tokens
$9.00
Cached input $ / 1M
$0.15
Each benchmark shown separately with its own source
Benchmark results (21)
| Benchmark | Domain | Score | vs best recorded | Setting | Run | Source |
|---|---|---|---|---|---|---|
| LMCA | agents | 47.1% | 69% | high | — | External ↗ |
| DTBench | reasoning | 94.7% | 96% | high | — | External ↗ |
| Mystery Game Puzzles | games | 32.0% ±4.7 | 38% | high | 27 Jul 2026 | Epoch ↗ |
| EBR-bench | reasoning | 4.8% | 6% | high | 25 Jun 2026 | Epoch ↗ |
| Surface Evolver Bench | science | 58.1% | 61% | medium | — | External ↗ |
| FrontierMath-Tiers-1-3-v2-Private | math | 62.8% ±2.9 | 67% | high | 10 Jun 2026 | Eval log ↗ |
| FrontierMath-Tier-4-v2-Private | math | 26.8% ±7.0 | 27% | high | 10 Jun 2026 | Eval log ↗ |
| DeepSWE | coding | 37.4% | 50% | medium | — | External ↗ |
| ProofBench | math | 31.0% | 31% | high | — | External ↗ |
| APEX-Agents | agents | 27.5% | 36% | unknown | — | External ↗ |
| Chess Puzzles | games | 50.0% ±5.0 | 69% | high | 28 May 2026 | Eval log ↗ |
| SimpleQA Verified | knowledge | 66.2% ±1.5 | 88% | high | 27 Aug 2026 | Eval log ↗ |
| FrontierMath-Tier-4-2025-07-01-Privatesuperseded | math | 14.6% ±5.1 | 30% | high | 25 May 2026 | Epoch ↗ |
| ARC-AGI-2 | reasoning | 72.1% | 76% | high | — | External ↗ |
| FrontierMath-2025-02-28-Privatesuperseded | math | 39.0% ±2.9 | 74% | high | 22 May 2026 | Epoch ↗ |
| WeirdML | coding | 62.6% ±0.0 | 67% | high | — | External ↗ |
| OTIS Mock AIME 2024-2025 | math | 95.6% ±2.7 | 96% | high | 25 May 2026 | Eval log ↗ |
| SimpleBench | reasoning | 76.7% | 94% | unknown | — | External ↗ |
| SWE-Bench verified | coding | 79.3% ±1.8 | 95% | high | 1 Jun 2026 | Eval log ↗ |
| GPQA diamond | science | 92.8% ±1.6 | 97% | high | 22 May 2026 | Eval log ↗ |
| ARC-AGI | reasoning | 92.5% | 94% | high | — | External ↗ |
Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.
API list price over time
Price history
$ per 1M tokens
5 recorded prices on 6 Jun 2026 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 40m ago. Steps show when the price changed.
Events
- BenchmarkEpoch AI evaluates Gemini 3.5 FlashEpoch AI Benchmarking Hub
- Model launchMajorGoogle releases Gemini 3.5 FlashEpoch AI Benchmarking Hub