Claude Opus 4.8
Released
Claude Opus 4.8 is an AI model released by Anthropic on Thursday, May 28 2026, 42 days after Claude Opus 4.7. It has a 1M token context window. Benchmark results (shown below) cover BullshitBench v2, BU Bench, Gray Swan IPI, ProgramBench, SWE-Bench Pro, SWE-Bench Verified, and 16 more.
Get Anthropic releases and AI news.
Every other Monday
API pricing
- InputWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it.
- $5.00
- Cached inputA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again.
- $0.50
- 5 min cache writeA one-off charge for storing text so later requests can reuse it at the cheaper cached rate. This option keeps it for five minutes.
- $6.25
- 1 hr cache writeA one-off charge for storing text so later requests can reuse it at the cheaper cached rate. This option keeps it for an hour, so it costs more than the five-minute one.
- $10.00
- OutputWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one.
- $25.00
Available from
| Amazon Bedrock | — | — | $5.00 | $25.00 | 1M |
|---|---|---|---|---|---|
| Anthropic | — | — | $5.00 | $25.00 | 1M |
| Azure | global | — | $5.00 | $25.00 | 1M |
| Claude Platform on AWS | — | — | $5.00 | $25.00 | 1M |
| global | — | $5.00 | $25.00 | 1M | |
| Amazon Bedrock | eu-west-1 | — | $5.50 | $27.50 | 1M |
| Amazon Bedrock | us | — | $5.50 | $27.50 | 1M |
| Azure | us | — | $5.50 | $27.50 | 1M |
| eu | — | $5.50 | $27.50 | 1M | |
| us | — | $5.50 | $27.50 | 1M | |
| Anthropic | — | fast | $10.00 | $50.00 | 1M |
Benchmarks
Coding
ProgramBenchProgram reconstruction — The AI receives a working program and its documentation, then builds a replacement from scratch without the original source code, internet access or decompilation. The score is the percentage of 200 programs that pass every behavioral test. We record each model's best published mini-SWE-agent result, including higher reasoning efforts where available. Partial test-pass rates and almost-solved programs do not count toward this score. Equal scores share a rank here; the official board also uses partial progress to break ties. Higher is better.via ProgramBench
0%
#6 of 20Best: Claude Opus 5 · 4.5%
SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better.
69.2%
#4 of 23Best: Claude Fable 5.1 · 81.2%
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better.via BenchLM
88.6%
#4 of 57Best: Claude Opus 5 · 96%
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better.
84.4%
#4 of 14Best: Claude Opus 5 · 89.5%
DeepSWE 1.0Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. Higher is better.
55.8%
#5 of 6Best: Kimi K3 · 67.5%
Next.js EvalsNext.js coding — Vercel'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.
68%
#17 of 32Best: Claude Fable 5.1 · 97%
Terminal & CLI
Frontier-Bench v0.1Agentic computer work — A 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.
21.1%
#4 of 9Best: Claude Opus 5 · 43.3%
Terminal-Bench 4.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Version 4.0 recalibrated how much time, CPU and memory each task gets, removed eight tasks and fixed nineteen, so fewer runs fail for reasons that have nothing to do with the model. Scores are not comparable with earlier versions. Higher is better.via Terminal-Bench
23.64%
#14 of 21Best: Claude Sonnet 5.5 · 70.6%
Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better.
74.6%
#23 of 32Best: DeepSeek-V4.1-Flash · 90.6%
Agentic & tool use
BU BenchBrowser agent — Can the AI drive a real web browser to finish tasks — clicking, filling forms, and navigating sites the way a person would? Run by Browser Use on their BU Bench task set. Higher is better.
74%
#1Best published BU Bench score of all tracked models
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better.via BenchLM
84.3%
#13 of 28Best: GPT-5.6 Sol · 92.2%
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better.
83.4%
#5 of 28Best: Qwen3.8-Max · 86.1%
Reasoning & science
Humanity's Last ExamMultidisciplinary reasoning — Humanity'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.
49.8%no tools
57.9%with tools
no tools
#3 of 22Best: Claude Fable 5.1 · 60.9%
with tools
#12 of 45Best: Claude Opus 5.5 · 67.7%
Knowledge work
GDPval-AAKnowledge work — Measures how well the AI does economically valuable knowledge work, judged against human experts. Shown as a rating (like a chess Elo) — higher is better.
1890
#2 of 9Best: Claude Fable 5 · 1932
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo)
1600
#10 of 20Best: Claude Opus 5 · 1861
Finance
Finance Agent v2Agentic financial analysis — Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better.
53.9%
#5 of 10Best: Gemini 4 Argon · 65.4%
Legal & tax
Harvey's Legal Agent BenchmarkAgentic legal work — Harvey's test of whether an AI agent can complete real legal work — drafting and reviewing documents, working with spreadsheets and presentations, and navigating files the way a lawyer's assistant would. Higher is better.
9.58%
#8 of 10Best: Muse Spark 1.1 · 20%
TaxEval v2Tax questions — A set of real tax questions created by Vals AI — can the AI give accurate answers about tax rules and filings? Higher is better.
75.63%
#3 of 4Best: Muse Spark 1.1 · 79.72%
Healthcare
MedScribeMedical admin work — Can the AI support doctors with their administrative work, like notes and paperwork? Created by Vals AI. Higher is better.
85.75%
#5 of 5Best: Muse Spark 1.1 · 88.89%
Robustness
BullshitBench v2Nonsense detection — Given 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.
95%
#1Best published BS-Bench score of all tracked models
Gray Swan IPIPrompt injection robustness — Attackers 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.
0.5%k = 1
4.1%k = 10
5.5%k = 15
k = 1
#3 of 13Best: Claude Opus 5 · 0.2%
k = 10
#3 of 13Best: Claude Opus 5 · 1.6%
k = 15
#4 of 14Best: Gemini 4 Argon · 0.7%
Source: ProgramBench, retrieved 11 September 2026 · Source: BenchLM, retrieved 31 August 2026 · Source: Terminal-Bench, retrieved 30 August 2026. Other sources are identified on the linked benchmark pages.
Compare Claude Opus 4.8 with
Suggested comparisons
Claude Opus 4.8vsClaude Sonnet 5.5Claude Opus 4.8vsGPT-6.1 SolClaude Opus 4.8vsGemini 4 ArgonClaude Opus 4.8vsMuse Spark 1.3Claude Opus 4.8vsGrok 4.7Claude Opus 4.8vsDeepSeek-V4.1-FlashClaude Opus 4.8vsMistral Large 4Claude Opus 4.8vsKimi K3Claude Opus 4.8vsGLM-5.3-FlashClaude Opus 4.8vsQwen3.8-Max-0902Claude Opus 4.8vsNemotron 3.5 LightningAbout
Claude Opus 4.8, released May 28, 2026, continued the rapid iteration of the Opus line with a 1M-token context window and top scores across agentic and real-world evaluations: 69.2% on SWE-Bench Pro, 84.4% on SWE-Bench Multilingual, and an Artificial Analysis GDPval Elo of 1890, the highest of any model at its release. Its 95% score on BullshitBench-V2 — a benchmark measuring whether a model pushes back on false premises — was also a record.
Opus 4.8 arrived just seven weeks after Opus 4.7 and held the flagship spot for less than two weeks before Claude Fable 5, the first of Anthropic's Claude 5 generation, launched on June 9, 2026. It also closed out the 4-series for the Opus name: Claude Opus 5 succeeded it on July 24, 2026 at the same price.
Frequently asked questions
Claude Opus 4.8 was released by Anthropic on Thursday, May 28 2026.
All Anthropic releases
29 tracked2026
12 releasesClaude Sonnet 5.5
Sep 28 2026
Claude Opus 5.5
Sep 22 2026
Claude Fable 5.1
Sep 1 2026
Claude Mythos 5.1
Sep 1 2026
Claude Opus 5
Jul 24 2026
Claude Sonnet 5
Jun 30 2026
Claude Fable 5
Jun 9 2026
Claude Mythos 5
Jun 9 2026
Claude Opus 4.8
May 28 2026
Claude Opus 4.7
Apr 16 2026
Claude Sonnet 4.6
Feb 17 2026
Claude Opus 4.6
Feb 5 2026