Claude Opus 5
An upcoming flagship model from Anthropic with capabilities approaching the previous generation's flagship level, but at significantly reduced inference cost. Its scheduled release is July 2025.
Core Facts
Anthropic称Opus 5是迄今最对齐的模型，遵守宪法程度超过Sonnet 5和Fable，欺骗行为率最低
Opus 5.5 reaches the level of Fable 5.1 on most tasks, but with approximately 40% lower operating costs and faster speed.
Anthropic states that Opus 5.5 can reduce total costs for typical tasks by approximately 40% and improve output speed by more than 30%.
Timeline (last 90 days)
Tev-27B 完全在 ALE（Agents' Last Exam）的计算与数学科学领域上使用 PPRS 训练，在 ALE-CM 的 Pass@5 指标上匹配了 GPT-6 Astra 和 Claude Opus 5。
Twenty minutes after the first report, the team updated the status to 'Identified'; at UTC 01:35 some models (Claude Fable 5 and 5.1, Mythos 5 and 5.1) recovered, at UTC 02:11 all affected models returned to normal and entered monitoring, and at UTC 02:35 the status was marked as 'Resolved.'
Anthropic's status page recorded an incident involving Claude Mythos 5.1 and Claude Fable 5.1, in which both models experienced intermittent increases in error rates during a similar time period.
在 Surge AI 组织的盲评中，专业标注员给 Mistral Large 4 预览版打出 3.74/5，在五个模型中排名第二，落后于 Claude Opus 5（4.22），领先于 GLM-5.3、Kimi K3 和 GLM-5.2。
On TerminalBench 2.1, HydraFusion reduces costs by 67% and achieves 4.9 points higher quality compared to Claude Opus 5.
On September 30, Anthropic's seat switched from Claude Opus 5 ($10.00) to Claude Opus 5.5 ($8.00), causing the index to fall back to $5.14.
To ensure fairness, all models had thinking disabled for direct answering; however, GPT-6 Astra does not support disabling thinking and could only be set to the lowest level 'low', so Claude Opus 5 was also set to 'low' for the head-to-head comparison.
GPT-6 Astra achieved the best performance among the eight models with an average score of 0.90, followed by Claude Opus 5 with an average score of 0.82.
In the test, the top tier (Overall 83.33) is occupied by Gemini 3.5 Flash, Claude Sonnet 5, and Claude Opus 5.
Claude Opus 5 frequently hit the 16,000-token limit due to excessively long reasoning, and a single full run costing approximately $19 exceeded Kaggle's $10-per-day per-model budget, so it only completed the first half, Part A.
40 more timeline events
All Facts (20)
Anthropic称Opus 5是迄今最对齐的模型，遵守宪法程度超过Sonnet 5和Fable，欺骗行为率最低
85%VerifiedOpus 5.5 reaches the level of Fable 5.1 on most tasks, but with approximately 40% lower operating costs and faster speed.
80%VerifiedAnthropic states that Opus 5.5 can reduce total costs for typical tasks by approximately 40% and improve output speed by more than 30%.
70%VerifiedAccounting for real usage scenarios, the actual cost advantage of Opus 5 over Fable is approximately 20% to 25%, with Fable costing about $2.75 per task and Opus 5 about $2.03.
65%VerifiedOpus 5's pass rate at Low Effort is approximately 60%, surpassing Opus 4.8's 59% at the Max level, for which the per-task cost is as high as $13.
65%VerifiedClaude 3.5 Sonnet has approached or even surpassed the performance of the early Claude 3 Opus in most benchmark tests.
65%VerifiedClaude Opus 5 is Anthropic's flagship large language model released in 2025, representing the highest capability tier of the Claude series
60%VerifiedAnthropic released Claude Sonnet 5, officially claiming its performance approaches that of Opus 4.8 at a lower price
60%VerifiedDevin has integrated Anthropic's Claude Opus 5 model, live in Devin Desktop and Devin CLI
65%Unverified能力提升型技能存在时效性问题，随着模型版本升级（如未来的Claude Opus 5），某些能力提升技能可能不再需要
70%UnverifiedIn the Automation Bench test, GPT-6 Sol's extra high tier achieved first place 33% of the time, while 5.6's max effort scored 28.8.
60%UnverifiedV4.1-Flash scores 1 point higher than a GPT-5.6-level model in comparison.
60%UnverifiedThe effort parameter controls how much the model thinks rather than how much it says; reducing effort does not reliably shorten visible responses, and controlling response length requires explicit prompting.
60%UnverifiedClaude Opus 5 scored 1861 on the knowledge work benchmark GDPval, above the human average.
60%UnverifiedClaude Opus 5 scored 30.2% on ARC-AGI-3, compared to only 1.5% for the previous generation.
60%UnverifiedOpus 5 remains proficient at discovering vulnerabilities, but its ability to exploit them has been significantly weakened.
60%UnverifiedClaude Opus 5 approaches Fable-level performance on the FrontierCode 1.1 benchmark while costing only half as much
60%UnverifiedAnthropic's status page recorded an incident involving Claude Mythos 5.1 and Claude Fable 5.1, in which both models experienced intermittent increases in error rates during a similar time period.
60%UnverifiedTev-27B 完全在 ALE（Agents' Last Exam）的计算与数学科学领域上使用 PPRS 训练，在 ALE-CM 的 Pass@5 指标上匹配了 GPT-6 Astra 和 Claude Opus 5。
50%Unverified在 Surge AI 组织的盲评中，专业标注员给 Mistral Large 4 预览版打出 3.74/5，在五个模型中排名第二，落后于 Claude Opus 5（4.22），领先于 GLM-5.3、Kimi K3 和 GLM-5.2。
50%