跳转到主要内容

Claude Mythos Preview

AnthropicClaudeProprietary

描述

Claude Mythos Preview is an unreleased general-purpose frontier model from Anthropic, a new tier above Opus (internal codename 'Capybara'). It identified thousands of zero-day vulnerabilities across every major operating system and web browser as part of Project Glasswing, a cross-industry cybersecurity initiative with 12 partners including AWS, Apple, Microsoft, and Google. State-of-the-art on SWE-bench Verified (93.9%), GPQA Diamond (94.6%), USAMO (97.6%), Terminal-Bench 2.0 (82.0%), CyberGym (83.1%), and Cybench (100% pass@1, saturated). Represents a 4.3x increase over the previous trendline for model performance. Deployed under ASL-3 Standard. Best-aligned Claude model to date per Anthropic's risk report, with the first-ever 24-hour internal alignment review before deployment. Not planned for general availability. Pricing for participants: $25/$125 per million tokens (input/output). 244-page system card.

发布日期
参数规模
上下文长度
支持模态
image, text

能力雷达图

90
general
80
coding
80
reasoning
77
science估算
80
agents
90
multimodal

Science 在缺少专门科学评测时使用推理能力代理估算。

排行榜排名

领域#排名分数来源
智能体与工具3
79.0
LS
多模态榜3
93.0
LS

基准测试分数 (LLM Stats)

Agents

CyBench100.0%自报
BrowseComp86.9%自报
CyberGym83.1%自报
Terminal-Bench 2.082.0%自报
OSWorld-Verified79.6%自报
SWE-Bench Pro77.8%自报
SWE-Bench Multimodal59.0%自报

Biology

GPQA94.6%自报

Code

SWE-Bench Verified93.9%自报
SWE-bench Multilingual87.3%自报

General

MMMLU92.7%自报

Healthcare

FigQA89.0%自报

Long Context

Graphwalks BFS >128k80.0%自报

Math

USAMO2597.6%自报
Humanity's Last Exam64.7%自报

Multimodal

CharXiv-R93.2%自报

AA 评测指数

暂无 AA 评测数据

LLM Stats 分类评分

Biology
90
Chemistry
90
Frontend Development
90
General
90
Healthcare
90
Language
90
Multimodal
90
Physics
90
Reasoning
90
Safety
90
Search
90
Spatial Reasoning
80
Tool Calling
80
Vision
80
Agents
80
Code
80
Long Context
80
Math
80

定价

暂无定价数据

速度

暂无速度数据

可用提供商

(LS 内部计价单位)

暂无提供商数据

外部链接