Compare / head-to-head
Gemini 3.6 FlashvsQwen3.5 397B-A17B
Gemini 3.6 Flash leads 4 of 4 shared benchmarks. Qwen3.5 397B-A17B is 3.7x cheaper per token. Gemini 3.6 Flash has the larger context window (1.0M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
4 – 0
Gemini 3.6 Flash leads
Cheaper per token
Qwen3.5 397B-A17B
3.7x cheaper, input + output
Larger context
Gemini 3.6 Flash
1.0M tokens
Provider uptime (30d)
100% · —
Google · Qwen
Gemini 3.6 Flash
Google · released 2026-07-21
textvisionaudio
Qwen3.5 397B-A17B
Qwen · released 2026-02-15
textvision
Quality
Benchmark matrix
4 shared · 10 only Gemini 3.6 Flash · 3 only Qwen3.5 397B-A17B| Benchmark | Gemini 3.6 Flash | Qwen3.5 397B-A17B | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 4 | ||||
| GPQA Diamond | 94.1% ↗epoch | 88.4% ↗ | +5.7 pt | Gemini 3.6 Flash |
| AIME 2026 | 96.7% ↗matharena ⚠ | 94.2% ↗matharena ⚠ | +2.5 pt | Gemini 3.6 Flash |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 58.9% ↗epoch | 29.5% ↗epoch | +29.4 pt | Gemini 3.6 Flash |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 94.2% ↗epoch | 88.9% ↗epoch | +5.3 pt | Gemini 3.6 Flash |
| Only Gemini 3.6 Flash reports · 10 | ||||
| CharXiv Reasoning (no tools) | 85.2% ↗ | not reported | — | — |
| DeepSWE v1.1 | 49% ↗ | not reported | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | 22% ↗epoch | not reported | — | — |
| GDM-MRCR v2 (128k average) | 91.8% ↗ | not reported | — | — |
| GDPVal-AA v2 | 1421 ↗ | not reported | — | — |
| MLE-Bench | 63.9% ↗ | not reported | — | — |
| OSWorld-Verified | 83% ↗ | not reported | — | — |
| SimpleQA Verified | 66.2% ↗epoch | not reported | — | — |
| SWE-Bench Pro (Public) | 58.7% ↗ | not reported | — | — |
| Terminal-bench 2.1 | 78% ↗ | not reported | — | — |
| Only Qwen3.5 397B-A17B reports · 3 | ||||
| SWE-bench Verified | not reported | 76.4% ↗ | — | — |
| MMLU-Pro | not reported | 87.8% ↗ | — | — |
| LMArena Elo | not reported | 1438.3 ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Gemini 3.6 Flash minus Qwen3.5 397B-A17B in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | Gemini 3.6 Flash | Qwen3.5 397B-A17B | Edge |
|---|---|---|---|
| Context window | 1.0M tokens | 262K tokens | Gemini 3.6 Flash |
| Max output | 66K tokens | — | — |
| Input price / 1M | $0.75 ↗ | $0.172 ↗ | Qwen3.5 397B-A17B |
| Output price / 1M | $3.75 ↗ | $1.03 ↗ | Qwen3.5 397B-A17B |
| Cached input / 1M | $0.075 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $4.5 | $1.2 | Qwen3.5 397B-A17B |
| Modalities | text · vision · audio | text · vision | Gemini 3.6 Flash |
| Released | 2026-07-21 | 2026-02-15 | — |
| Cited benchmark scores | 14 | 8 | — |
Reliability
Provider status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
Gemini 3.6 Flash vs …
Models sharing the most benchmarksMore matchups
Qwen3.5 397B-A17B vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Gemini 3.6 Flash and Qwen3.5 397B-A17B on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.