Gemini 3.6 Flash vs Gemini 3.8 Flash
| Spec | Gemini 3.6 Flash | Gemini 3.8 Flash |
|---|---|---|
| Respan Index | 44.4 (#48) | 54.6 (#22) |
| Blended price per 1M | $0.75 | $0.75 |
| Context window | 1.0M | 1.0M |
| Input / output per 1M | $0.375 / $1.88 | $0.375 / $1.88 |
| Max output | 66K | 66K |
| Input | text, vision, audio | text, vision, audio |
| Released | 2026-07-21 | 2026-09-02 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | Gemini 3.6 Flash | Gemini 3.8 Flash |
|---|---|---|
| Coding | 18.1 Gemini 3.6 Flash | 36.7 Gemini 3.8 Flash |
| Agents | 57.3 Gemini 3.6 Flash | 66.2 Gemini 3.8 Flash |
| Reasoning | 43.4 Gemini 3.6 Flash | 53.7 Gemini 3.8 Flash |
| Knowledge & language | 65.3 Gemini 3.6 Flash | 70.9 Gemini 3.8 Flash |
| Human preference | 54.5 Gemini 3.6 Flash | 55.8 Gemini 3.8 Flash |
Benchmarks both report
Gemini 3.8 Flash leads on 23 of the 27 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Gemini 3.6 Flash | Gemini 3.8 Flash |
|---|---|---|
| Chess Puzzles (Epoch AI) | 42% | 58% |
| LMArena Elo | 1483 | 1496 |
| LMArena WebDev (LMArena) | 1538 | 1583 |
| LMArena Vision (LMArena) | 1282 | 1291 |
| LMArena Agent (LMArena) | -0.0445 | 0.0265 |
| APEX-Agents (Mercor) | 46.9% | 64.3% |
| DeepSWE v1.1 (Datacurve) | 46.7% | 73.8% |
| FrontierMath Tier 4 v2 (Epoch AI) | 22% | 22% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 58.9% | 68.4% |
| Furniture Assembly (Epoch AI) | 23.3% | 31.7% |
| GPQA Diamond | 94.1% | 95.4% |
| LiveBench Agentic Coding (LiveBench) | 43.4% | 54.2% |
| LiveBench Coding (LiveBench) | 77.9% | 72.5% |
| LiveBench Data Analysis (LiveBench) | 63% | 54% |
| LiveBench Instruction Following (LiveBench) | 75.4% | 81.4% |
| LiveBench Language (LiveBench) | 83.9% | 87.8% |
| LiveBench Mathematics (LiveBench) | 86.4% | 91.6% |
| LiveBench Reasoning (LiveBench) | 85.1% | 89.3% |
| Mystery Game Puzzles (Epoch AI) | 30% | 47% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 94.2% | 98.9% |
| SAGE (Vals AI) | 48.7% | 35.1% |
| SimpleQA Verified | 66.2% | 69.7% |
| WeirdML (Håvard Tveit Ihle) | 56.1% | 84.8% |
| CharXiv Reasoning (no tools) | 85.2% | 86.2% |
| DeepSWE v1.1 | 49% | 73.7% |
| GDPVal-AA v2 | 1421 | 1545 |
| Terminal-bench 2.1 | 78% | 89.4% |
FAQ
Is Gemini 3.6 Flash better than Gemini 3.8 Flash?
Gemini 3.8 Flash scores 54.6 on the Respan Index (#22), ahead of Gemini 3.6 Flash at 44.4 (#48). Gemini 3.8 Flash leads on 23 of the 27 benchmarks both report.
Which is cheaper, Gemini 3.6 Flash or Gemini 3.8 Flash?
Gemini 3.6 Flash lists at $0.375 input and $1.88 output per 1M tokens; Gemini 3.8 Flash at $0.375 and $1.88. Both cost about the same per token at list price.
Which has the larger context window?
Both accept 1.0M tokens of context.