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。
Claude Mythos Preview
Anthropic
Gemini 3.1 Pro Preview
Google Deep Mind
Each axis is a category average, normalized to a 100-point radar.
Relative edge: 编程与软件工程 +18.4 / Relative gap: AI Agent - 信息收集 -1.0
Relative edge: AI Agent - 信息收集 +1.0 / Relative gap: 编程与软件工程 -18.4
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
Claude Mythos Preview · 202.78
Best single
Claude Mythos Preview · METR Time Horizons v1.1 1044.78
Modality coverage
Gemini 3.1 Pro Preview · 4 modalities
Head to head
8
Benchmarks
7
Wins
1
Losses
+90.88
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.
8 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | Claude Mythos Preview | Gemini 3.1 Pro Preview |
|---|---|---|
HLE 综合评估 | 64.70Extended Thinking ｜ Tools | 51.40Thinking Level · High ｜ Tools |
GPQA Diamond 科学与综合推理 | 94.60Extended Thinking | 94.30Thinking Level · High |
SWE-Bench Pro - Public 编程与软件工程 | 77.80Extended Thinking ｜ Tools | 54.20Thinking Level · High ｜ Tools |
SWE-bench Verified 编程与软件工程 | 93.90Extended Thinking ｜ Tools | 80.60Thinking Level · High ｜ Tools |
BrowseComp AI Agent - 信息收集 | 84.90Extended Thinking ｜ Tools | 85.90Thinking Level · High ｜ Tools |
OSWorld-Verified AI Agent - 工具使用 | 79.60Extended Thinking ｜ Tools | 76.20Thinking Enabled ｜ Tools |
Terminal Bench 2.0 AI Agent - 工具使用 | 82.00Extended Thinking ｜ Tools | 68.50Thinking Level · High ｜ Tools |
METR Time Horizons v1.1 Agent能力评测 | 1044.78Standard Mode ｜ Tools | 384.15Standard Mode ｜ Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | Claude Mythos PreviewAnthropic | Gemini 3.1 Pro PreviewGoogle Deep Mind |
|---|---|---|
Core specsRelease | 2026-04-07 | 2026-02-20 |
Context length | — | 1M |
Max output | 8192 | 65536 |
MoE | No | No |
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 | Not provided | / |
Video Input/Output | Not provided | / |
ResourcesPaper / report | Introducing Claude Mythos Preview and Project Glasswing | Gemini 3.1 Pro: A smarter model for your most complex tasks |
DataLearner blog | Claude Mythos 是什么？Anthropic最强模型评测、安全能力与Project Glasswing详解 | Not provided |
