Claude Sonnet 4.6vsGemini 3.0 Pro (Preview 11-2025)
Across 11 shared benchmarks, Claude Sonnet 4.6 leads overall: Claude Sonnet 4.6 wins 7, Gemini 3.0 Pro (Preview 11-2025) wins 3, with 1 ties and an average score difference of +5.66.
Anthropic · 2026-02-17 · Chat model
Google Deep Mind · 2025-11-18 · Multimodal model
Benchmark scores
Grouped by capability, sorted by largest gap within each. 11 shared benchmarks.
General Knowledge
Claude Sonnet 4.6 2/3| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| ARC-AGI-2 | 58.3018 / 59 | 45.1023 / 59 | +13.20 |
| GPQA Diamond | 89.9021 / 178 | 93.805 / 178 | -3.90 |
| HLE | 4925 / 157 | 45.8031 / 157 | +3.20 |
Agent Level Benchmark
Gemini 3.0 Pro (Preview 11-2025) 1/1| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| τ²-Bench - Telecom | 97.909 / 35 | 985 / 35 | -0.10 |
AI Agent - Information Search
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| BrowseComp | 74.7020 / 45 | 59.2031 / 45 | +15.50 |
AI Agent - Tool Usage
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| Terminal Bench 2.0 | 59.1022 / 46 | 56.9025 / 46 | +2.20 |
Claw-style Agent Evaluation
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| Pinch Bench | 885 / 37Thinking (With Tools) | 70.7031 / 37Thinking (With Tools) | +17.30 |
Coding and Software Engineer
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| SWE-bench Verified | 79.6017 / 108 | 76.2032 / 108 | +3.40 |
Long Context
Even 1/1| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| AA-LCR | 711 / 13 | 711 / 13 | — |
Math and Reasoning
Gemini 3.0 Pro (Preview 11-2025) 1/1| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 8.3034 / 80Thinking (No Tools, 16K Budget) | 18.8016 / 80 | -10.50 |
Productivity Knowledge
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) | Diff |
|---|---|---|---|
| GDPval-AA | 5711 / 21 | 3518 / 21 | +22 |
Specs
| Field | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) |
|---|---|---|
| Publisher | Anthropic | Google Deep Mind |
| Release date | 2026-02-17 | 2025-11-18 |
| Model type | Chat model | Multimodal model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 1000K |
| Max output | 8K | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Claude Sonnet 4.6 | Gemini 3.0 Pro (Preview 11-2025) |
|---|---|---|
| Text input | $3 / 1M tokens | Not public |
| Text output | $15 / 1M tokens | Not public |
| Cache read | $0.3 / 1M tokens | Not public |
| Cache write | $3.75 / 1M tokens | Not public |
One or both models have incomplete public pricing.
Summary
- Claude Sonnet 4.6leads in:General Knowledge (2/3), AI Agent - Information Search (1/1), AI Agent - Tool Usage (1/1), Claw-style Agent Evaluation (1/1), Coding and Software Engineer (1/1), Productivity Knowledge (1/1)
- Gemini 3.0 Pro (Preview 11-2025)leads in:Agent Level Benchmark (1/1), Math and Reasoning (1/1)
- Tied in:Long Context
On average across the 11 shared benchmarks, Claude Sonnet 4.6 scores 5.66 higher.
Largest single-benchmark gap: GDPval-AA — Claude Sonnet 4.6 57 vs Gemini 3.0 Pro (Preview 11-2025) 35 (+22).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.