Gemini 3.8 Flash
Gemini 3.8 Flash is an API-only (closed weights) language model from Google, released September 2, 2026. The context window is 1M tokens. On the Epoch Capabilities Index it scores 158.8, rank 23 of 256 rated models. API price: $0.75 input / $3.75 output per million tokens.
Gemini 3.8 Flash is our most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows—all with the speed and cost efficiency of Flash.
Check it against your own PC →
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Reasoning | 37.9 | Chess Puzzles, SimpleBench, Mystery Game Puzzles, Humanity's Last Exam |
| Math | 51.6 | ProofBench, OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private |
| Coding | 58.8 | MirrorCode, FrontierSWE, DeepSWE |
| Knowledge | 82.8 | GPQA diamond, SimpleQA Verified |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| OTIS Mock AIME | 98.9% |
| GPQA Diamond | 97.8% |
| SimpleBench | 78.9% |
| DeepSWE | 73.8% |
| FrontierMath (Tiers 1–3) | 72.0% |
| SimpleQA Verified | 69.7% |
| Chess Puzzles | 55.8% |
| ProofBench | 48.0% |
| Humanity's Last Exam | 44.5% |
| Mystery Game Puzzles | 41.6% |
| MirrorCode | 38.9% |
| FrontierMath-Tier-4-v2-Private | 22.0% |
| FrontierSWE | 19.6% |
| Remote Labor Index | 5.8% |
| Furniture Assembly | 2.4% |
Links
Announcement · Epoch AI benchmarks
Models with a similar score
GPT-5.5 Pro · Muse Wonder · Kimi K3.1 Pro · GPT-5.6 Terra · Claude Haiku 5.5 · GPT-5.4 Pro · GPT-5.5 · Claude Opus 4.8
More from Google
Gemini 4 Carbon · Gemini 4 Argon · Gemini 3.7 Flash · Gemini 3.6 Flash · Gemini 3.5 Flash-Lite · Gemini 3.5 Flash Cyber · Gemma 4 E4B · Gemma 4 E2B