跳轉到主要內容

Claude Opus 4.6 (Non-reasoning, High)

AnthropicClaudeProprietary

描述

Claude Opus 4.6 is Anthropic's most intelligent model, improving on its predecessor's coding skills with more careful planning, longer agentic task sustenance, more reliable operation in larger codebases, and better code review and debugging skills. First Opus-class model with 1M token context window (beta), 128K output tokens, and adaptive thinking. Features effort controls (low/medium/high/max) and context compaction for long-running tasks. State-of-the-art on Terminal-Bench 2.0, Humanity's Last Exam, GDPval-AA, and BrowseComp. Pricing: $5/$25 per million tokens (input/output).

發布日期
2026-02-05
參數規模
—
上下文長度
1.0M
支援模態
image, pdf, text

能力雷達圖

100
general
70
coding
80
reasoning
77
science估算
80
agents
80
multimodal

缺少專門科學評測時,Science 由 LLM Stats 科學得分或推理能力估算。

排行榜排名

領域#排名分數來源
智慧體能力模型榜41
57.0
LS
多模態榜25
61.0
LS

基準測試分數 (LLM Stats)

(LLM Stats (zeroeval))

Agents

Finance Agent60.7%自報

Code

CyberGym73.8%自報
FrontierSWE56.0%
OpenRCA34.9%自報

Communication

Tau2 Telecom99.3%自報
Tau2 Retail91.9%自報

Language

MMMLU91.1%自報

Legal

Legal Agent Benchmark4.2%

Long Context

MRCR v2 (8-needle)76.0%自報

Math

AIME 202599.8%自報
LiveBench76.3%

Multimodal

OSWorld72.7%自報

Reasoning

Vending-Bench 2801759.0%自報
Graphwalks parents >128k95.4%自報
GPQANYU + Cohere + Anthropic (2023)91.3%自報
BrowseCompOpenAI (2025)84.0%自報
SWE-Bench Verified80.8%自報
SWE-bench Multilingual77.8%自報
CharXiv-R77.4%自報
ARC-AGI v268.8%自報
Terminal-Bench 2.0Stanford × Laude Institute (2026)65.4%自報
MCP Atlas62.7%自報
Graphwalks BFS >128k61.5%自報
Humanity's Last Exam53.1%自報

Safety

FigQA78.3%自報

Search

DeepSearchQA91.3%自報

Vision

MMMU-Pro77.3%自報

AA 評測指數

(Artificial Analysis)

暫無 AA 評測資料

LLM Stats 分類評分

(LLM Stats (zeroeval))
Agents
100
Reasoning
100
General
100
Communication
100
Language
90
Physics
90
Search
90
Biology
90
Chemistry
90
Long Context
80
Math
80
Multimodal
80
Safety
80
Spatial Reasoning
80
Frontend Development
80
Healthcare
80
Tool Calling
80
Code
70
Vision
70
Finance
60
Legal
0

定價

輸入價格$5 / 1M tokens
輸出價格$25 / 1M tokens
混合價格(3:1)$10 / 1M tokens
快取讀取價格$0.5 / 1M tokens
快取寫入價格$6.25 / 1M tokens

速度

Tokens/秒38.7
首Token延遲1.74s
首回答延遲1.74s

供應商價格排行

供應商價格排行

1 個供應商

供應商輸入輸出
1Anthropic主要
$5
$25

比較該模型在不同 API 供應商之間的定價。

外部連結