See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 2 个模型的评测数据与核心参数。
Gemini 3.0 Pro (Preview 11-2025)
Google Deep Mind
Gemini 3.1 Pro Preview
Google Deep Mind
Each axis is a category average, normalized to a 100-point radar.
Relative edge: 数学推理 +1.6 / Relative gap: 综合评估 -17.0
Relative edge: 综合评估 +17.0 / Relative gap: 数学推理 -1.6
Method: for each model and benchmark, the chart first averages all scores in the current mode scope instead of taking the best score, then averages those benchmark scores within each category. Only benchmarks with at least two selected models scored are included; missing values are not counted as zero.
Best overall
Gemini 3.1 Pro Preview · 74.51
Best single
Gemini 3.1 Pro Preview · τ²-Bench - Telecom 99.30
Modality coverage
Gemini 3.0 Pro (Preview 11-2025) · 4 modalities
Head to head
15
Benchmarks
3
Wins
12
Losses
-8.90
Average diff
Compare benchmark results across thinking modes and tool usage.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Complete scores for each model/mode across selected benchmarks.
15 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | Gemini 3.0 Pro (Preview 11-2025) | Gemini 3.1 Pro Preview |
|---|---|---|
ARC-AGI-2 综合评估 | 31.10Thinking Enabled | 77.10Thinking Level · High |
GPQA Diamond 综合评估 | 91.90Thinking Enabled | 94.30Thinking Level · High |
HLE 综合评估 | 45.80Thinking Level · High ｜ Tools | 51.40Thinking Level · High ｜ Tools |
LiveBench 综合评估 | 73.39Thinking Level · High | 79.93Thinking Level · High |
LiveCodeBench 编程与软件工程 | 92.00Thinking Enabled | 91.70Thinking Level · High ｜ Tools |
SWE-bench Verified 编程与软件工程 | 76.20Thinking Enabled | 80.60Thinking Level · High ｜ Tools |
FrontierMath 数学推理 | 38.00Thinking Enabled | 36.90Thinking Level · High |
18.80Thinking Enabled | 16.70Standard Mode | |
Simple Bench 常识推理 | 76.40Thinking Enabled | 79.60Standard Mode |
τ²-Bench Agent能力评测 | 85.40Thinking Enabled ｜ Tools | 90.80Thinking Level · High ｜ Tools |
τ²-Bench - Telecom Agent能力评测 | 98.00Thinking Level · High ｜ Tools | 99.30Thinking Level · High ｜ Tools |
BrowseComp AI Agent - 信息收集 | 59.20Thinking Level · High ｜ Tools | 85.90Thinking Level · High ｜ Tools |
MCP-Atlas AI Agent - 工具使用 | 70.30Standard Mode ｜ Tools | 78.20Thinking Level · High ｜ Tools |
Terminal Bench 2.0 AI Agent - 工具使用 | 56.90Thinking Level · High ｜ Tools | 68.50Thinking Level · High ｜ Tools |
Pinch Bench OpenClaw智能体能力综合测评 | 70.70Thinking Enabled ｜ Tools | 86.70Thinking Enabled ｜ Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | Gemini 3.0 Pro (Preview 11-2025)Google Deep Mind | Gemini 3.1 Pro PreviewGoogle Deep Mind |
|---|---|---|
Core specsRelease | 2025-11-18 | 2026-02-20 |
Context length | 1000K | 1M |
Max output | 65536 | 65536 |
MoE | No | No |
Supported modes | 常规模式（Non-Thinking Mode）思考模式（Thinking Mode）深度思考（Deeper Thinking Mode） | No mode data |
LicenseCode Open Source | Not provided | Not provided |
Weights Open Source | Not provided | Not provided |
Commercial use | 不开源 | 不开源 |
Modality supportText Input/Output | / | / |
Image Input/Output | / | / |
Audio Input/Output | / | / |
Video Input/Output | / | / |
ResourcesPaper / report | A new era of intelligence with Gemini 3 | Gemini 3.1 Pro: A smarter model for your most complex tasks |
DataLearner blog | 大模型新王者！谷歌发布Gemini 3.0 Pro，各方面评测几乎都是第一，全球首个大模型匿名投票得分超1500分的模型，支持100万输入上下文！ | Not provided |
