Claude Opus 5
Claude Opus 5 is an AI model released by Anthropic on Friday, Jul 24 2026, 24 days after Claude Sonnet 5. It has a 1M token context window. Benchmark results (shown below) cover Gray Swan IPI, FrontierCode v1.1 (Main), Frontier-Bench v0.1, Humanity's Last Exam, ARC-AGI-3, BioMysteryBench, and 13 more.
Benchmarks
Prompt injection robustness
Gray Swan IPIAttackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. Gray Swan's indirect prompt injection benchmark measures how often such an attack succeeds when the attacker gets a single try. Lower is better.
0.2%
k = 1
1.6%
k = 10
2%
k = 15
k = 1
#1Best published Gray Swan score of all tracked models
k = 10
#1Best published Gray Swan score of all tracked models
k = 15
#1Best published Gray Swan score of all tracked models
Agentic coding
FrontierCode v1.1 (Main)A set of very hard, frontier-difficulty coding tasks an AI agent has to complete end to end. The score is the share of tasks in the main split it solves. Higher is better.
53.4%
Agentic computer work
Frontier-Bench v0.1A hard, ever-evolving set of real computer tasks — coding, system administration, data work, and more — that an AI agent has to complete on its own. Run by the Harbor / Laude Institute team as the successor to Terminal-Bench (v0.1 is the first release of the task set). The score is the share of tasks solved. Higher is better.
43.3%
#1Best published Frontier v0.1 score of all tracked models
Multidisciplinary reasoning
Humanity's Last ExamHumanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better.
56.3%
no tools
64.7%
with tools
no tools
#1Best published HLE score of all tracked models
with tools
#1Best published HLE score of all tracked models
Novel problem-solving
ARC-AGI-3The third generation of the ARC-AGI series: instead of static puzzles, the AI is dropped into small interactive game-like environments it has never seen and has to figure out the rules and solve them on its own. Higher is better.
30.2%
Biology
BioMysteryBenchReal unsolved-style biology puzzles — the AI has to reason its way to an answer the way a research biologist would. The “hard” split contains the toughest cases. Higher is better.
49.4%
hard
90.1%
human solved
hard
human solved
Agentic computer use
OSWorld 2.0Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Version 2.0 is a harder, refreshed task set. Higher is better.
70.6%
Business workflows
AutomationBenchTests whether the AI can run real multi-step business workflows — the kind of end-to-end office processes companies want to automate — from start to finish. Higher is better.
26%
#1Best published AutoBench score of all tracked models
Agentic legal work
Harvey's Legal Agent Benchmark (Held-out)Harvey's test of whether an AI agent can complete real legal work, scored on a held-out set of tasks the model makers never see — making the numbers harder to game. Higher is better.
11.7%
Health
HealthBench ProfessionalRealistic health conversations graded against detailed rubrics written by physicians — can the AI respond the way a careful medical professional would? Higher is better.
59.8%
Knowledge work
GDPval-AA v2economically valuable knowledge work (v2, re-based Elo)
1861
#1Best published GDPval v2 score of all tracked models
Nonsense detection
BullshitBench v2Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better.
73%
#9 of 67Best: Claude Opus 4.8 · 95%
Agentic coding
CursorBench v3.2Cursor's own test of harder, real-world coding tasks inside a code editor, on the refreshed v3.2 task set. Scores aren't comparable with v3.1. Higher is better.
70%
#2 of 14Best: Claude Fable 5 · 70.5%
Agentic coding
DeepSWE 1.1Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better.
68.8%
#4 of 14Best: GPT-5.6 Sol · 73%
Next.js coding
Next.js EvalsVercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better.
88%
#5 of 24Best: Composer 2.5 · 92%
Supabase coding
Supabase EvalsSupabase's own open benchmark: a coding agent is dropped into a real Supabase project and asked to do real work — set up a schema, fix a broken security policy, debug an Edge Function — and every run is checked against a live Supabase stack. This is the headline number, where the agent has Supabase's own skills loaded, as most people building on Supabase would. The score is the share of scenarios it got right. Higher is better.
94.7%
with skills
94.7%
no skills
with skills
#2 of 5Best: GPT-5.6 Sol · 100%
no skills
#2 of 5Best: Kimi K3 · 100%
Web browsing
BrowseCompCan the AI browse the web and track down hard-to-find answers? Higher is better.
90.8%
#2 of 17Best: Kimi K3 · 91.2%
Community preference
Arena Elo (Text)Real people chat with two anonymous AIs side by side and vote for the answer they prefer. Votes become a chess-style Elo rating on arena.ai — it measures which AI people actually like, not test scores. Higher is better.
1495
#4 of 17Best: Claude Fable 5 · 1509
Community preference (code)
Arena Elo (Code)Like the text arena, but people vote on which AI writes better code. The votes become a chess-style Elo rating on arena.ai. Higher is better.
1673
#2 of 43Best: Kimi K3 · 1679
About Claude Opus 5
Claude Opus 5, released July 24, 2026, brought Anthropic's Claude 5 generation to the Opus tier six weeks after Claude Fable 5 opened it — at half Fable's price, keeping Opus 4.8's $5 per million input tokens and $25 per million output. The pitch was flagship-class agentic capability at workhorse pricing: at launch it scored 43.3% on Frontier-Bench v0.1, more than double Opus 4.8's 21.1% and nearly ten points clear of Fable 5, and posted a GDPval-AA v2 Elo of 1861 for knowledge work, the best published score at the time.
The launch card leaned on breadth: 90.8% on BrowseComp for agentic search, 70.6% on OSWorld 2.0 computer use, 64.7% on Humanity's Last Exam with tools, and 30.2% on ARC-AGI-3 — which Anthropic reported as roughly three times the next best published result on the novel problem-solving benchmark. Anthropic made it the default model on Claude Max and the strongest model available on Claude Pro, positioning Opus 5 as the everyday frontier model while Fable 5 kept the edge on a handful of evaluations, including DeepSWE agentic coding and Harvey's held-out legal benchmark.
Claude Opus 5 — frequently asked questions
- When was Claude Opus 5 released?
- Claude Opus 5 was released by Anthropic on Friday, Jul 24 2026.
- Who made Claude Opus 5?
- Claude Opus 5 was built by Anthropic. AI safety company building the Claude family of models. Founded in 2021 by former OpenAI researchers.
- What benchmark scores did Claude Opus 5 get?
- Claude Opus 5 reports 24 tracked benchmark scores — BullshitBench v2: 73%; Gray Swan IPI (k = 1): 0.2%; Gray Swan IPI (k = 10): 1.6%; Gray Swan IPI (k = 15): 2%; CursorBench v3.2: 70%; DeepSWE 1.1: 68.8%; FrontierCode v1.1 (Main): 53.4%; Next.js Evals: 88%; Supabase Evals (with skills): 94.7%; Supabase Evals (no skills): 94.7%; Frontier-Bench v0.1: 43.3%; BrowseComp: 90.8%; Humanity's Last Exam (no tools): 56.3%; Humanity's Last Exam (with tools): 64.7%; ARC-AGI-3: 30.2%; BioMysteryBench (hard): 49.4%; BioMysteryBench (human solved): 90.1%; OSWorld 2.0: 70.6%; AutomationBench: 26%; Harvey's Legal Agent Benchmark (Held-out): 11.7%; HealthBench Professional: 59.8%; GDPval-AA v2: 1861; Arena Elo (Text): 1495; Arena Elo (Code): 1673. Scores are the figures published at release by Anthropic. It holds the best score among all models tracked here on Gray Swan IPI (k = 1), Gray Swan IPI (k = 10), Gray Swan IPI (k = 15), FrontierCode v1.1 (Main), Frontier-Bench v0.1, Humanity's Last Exam (no tools), Humanity's Last Exam (with tools), ARC-AGI-3, BioMysteryBench (hard), BioMysteryBench (human solved), OSWorld 2.0, AutomationBench, Harvey's Legal Agent Benchmark (Held-out), HealthBench Professional and GDPval-AA v2.
- What is the context window of Claude Opus 5?
- Claude Opus 5 has a context window of 1M. That is the maximum amount of input plus output the model can hold in a single request.
- Is Claude Opus 5 open source?
- No. Claude Opus 5 is a proprietary model. The weights are not published — it is available only through the provider's own API, apps, or partner platforms.
- What came before and after Claude Opus 5?
- Anthropic's previous tracked release was Claude Sonnet 5 on Jun 30 2026, 24 days earlier. It is the most recent Anthropic model tracked on AI Release Tracker.
Compare Claude Opus 5 with
Claude Opus 5vsClaude Sonnet 5Claude Opus 5vsGPT-5.6 SolClaude Opus 5vsGemini 3.6 FlashClaude Opus 5vsMuse Spark 1.2Claude Opus 5vsGrok 4.5Claude Opus 5vsDeepSeek-V4-Flash-0731Claude Opus 5vsMistral Medium 3.5Claude Opus 5vsKimi K3Claude Opus 5vsComposer 2.5Claude Opus 5vsGLM-5.2Claude Opus 5vsQwen3.8-Max