Gemini 3.6 FlashvsClaude Sonnet 5
Across 9 shared benchmarks, Claude Sonnet 5 leads overall: Gemini 3.6 Flash wins 3, Claude Sonnet 5 wins 6, with 0 ties and an average score difference of -23.04.
Gemini 3.6 Flash
Google Deep Mind · 2026-07-21 · Multimodal model
Claude Sonnet 5
Anthropic · 2026-06-30 · Multimodal model
Gemini 3.6 Flash3 wins(33%)(67%)6 winsClaude Sonnet 5
Benchmark scores
Grouped by capability, sorted by largest gap within each. 9 shared benchmarks.
AI Agent - Tool Usage
Even 2/2| Benchmark | Gemini 3.6 Flash | Claude Sonnet 5 | Diff |
|---|---|---|---|
| Terminal-Bench 2.1 | 7827 / 49Thinking (With Tools) | 80.4023 / 49Extra-High (With Tools) | -2.40 |
| OSWorld-Verified | 835 / 26Thinking (With Tools) | 81.206 / 26Extra-High (With Tools) | +1.80 |
Coding and Software Engineer
Claude Sonnet 5 2/2| Benchmark | Gemini 3.6 Flash | Claude Sonnet 5 | Diff |
|---|---|---|---|
| WeirdML v2 | 56.1033 / 52Thinking High (With Tools) | 68.7820 / 52Thinking High (With Tools) | -12.68 |
| DeepSWE | 4928 / 35Thinking (With Tools) | 5424 / 35Deep Thinking (With Tools) | -5 |
Math and Reasoning
Claude Sonnet 5 2/2| Benchmark | Gemini 3.6 Flash | Claude Sonnet 5 | Diff |
|---|---|---|---|
| FrontierMath Tier 4 v2 | 21.9525 / 41Thinking High (No Tools) | 29.2719 / 41Max (No Tools) | -7.32 |
| FrontierMath v2 | 58.9523 / 58Thinking High (No Tools) | 65.6120 / 58Max (No Tools) | -6.67 |
General Evaluation
Gemini 3.6 Flash 1/1| Benchmark | Gemini 3.6 Flash | Claude Sonnet 5 | Diff |
|---|---|---|---|
| GPQA Diamond | 94.138 / 271Thinking High (No Tools) | 90.5340 / 271Extra-High (No Tools) | +3.60 |
Text Embedding
Gemini 3.6 Flash 1/1| Benchmark | Gemini 3.6 Flash | Claude Sonnet 5 | Diff |
|---|---|---|---|
| Context Arena | 88.7817 / 126Thinking High (No Tools) | 79.5340 / 126Max (No Tools) | +9.25 |
Writing and Creative Capabilities
Claude Sonnet 5 1/1| Benchmark | Gemini 3.6 Flash | Claude Sonnet 5 | Diff |
|---|---|---|---|
| Creative Writing | 1,60034 / 99Normal (No Tools) | 1,78818 / 99Normal (No Tools) | -187.90 |
Specs
| Field | Gemini 3.6 Flash | Claude Sonnet 5 |
|---|---|---|
| Publisher | Google Deep Mind | Anthropic |
| Release date | 2026-07-21 | 2026-06-30 |
| Model type | Multimodal model | Multimodal model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 1M |
| Max output | 64K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Gemini 3.6 Flash | Claude Sonnet 5 |
|---|---|---|
| Text input | $1.5 / 1M tokens | $2 / 1M tokens |
| Text output | $7.5 / 1M tokens | $10 / 1M tokens |
| Cache read | $0.15 / 1M tokens | $0.2 / 1M tokens |
| Cache write | Not public | $2.5 / 1M tokens |
Summary
- Gemini 3.6 Flashleads in:General Evaluation (1/1), Text Embedding (1/1)
- Claude Sonnet 5leads in:Coding and Software Engineer (2/2), Math and Reasoning (2/2), Writing and Creative Capabilities (1/1)
- Tied in:AI Agent - Tool Usage
On average across the 9 shared benchmarks, Claude Sonnet 5 scores 23.04 higher.
Largest single-benchmark gap: Creative Writing — Gemini 3.6 Flash 1,600 vs Claude Sonnet 5 1,788 (-187.90).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.