Claude Opus 4.8vsGPT-5.6 Terra
Claude Opus 4.8 | GPT-5.6 Terra | |
|---|---|---|
| Specifications | ||
Context window | 1M | — |
| Benchmarks | ||
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. | 95%Best | 53% |
Prompt injection robustness Gray Swan IPI · k = 1Attackers 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% | 5.4%Best |
Prompt injection robustness Gray Swan IPI · k = 10Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 10 tries and counts an attack as successful if any of them works. Lower is better. | 4.1% | 26%Best |
Prompt injection robustness Gray Swan IPI · k = 15Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 15 tries and counts an attack as successful if any of them works. Lower is better. | 5.5% | 30.4%Best |
Agentic coding SWE-Bench ProCan 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% | — |
Multilingual coding SWE-Bench MultilingualLike 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% | — |
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. | 62.3% | 64.9%Best |
Agentic coding CursorBench v3.1Cursor's own test of harder, real-world coding tasks inside a code editor. Higher is better. | 63.8% | — |
Agentic coding DeepSWE 1.0Artificial 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% | — |
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% | — |
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. | 21.1%Best | 20.8% |
Agentic terminal coding Terminal-Bench 2.1Can 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% | 84.3%Best |
Browser agent BU BenchCan 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% | — |
Web browsing BrowseCompCan the AI browse the web and track down hard-to-find answers? Higher is better. | 84.3% | — |
Multidisciplinary reasoning Humanity's Last Exam · no toolsHumanity'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% | — |
Multidisciplinary reasoning Humanity's Last Exam · with toolsHumanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | 57.9% | — |
Agentic computer use OSWorld-VerifiedCan the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. | 83.4% | — |
Agentic financial analysis Finance Agent v2Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better. | 53.9% | — |
Agentic legal work Harvey's Legal Agent BenchmarkHarvey'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% | — |
Tax questions TaxEval v2A 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% | — |
Medical admin work MedScribeCan the AI support doctors with their administrative work, like notes and paperwork? Created by Vals AI. Higher is better. | 85.75% | — |
Knowledge work GDPval-AAMeasures 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 | — |
Knowledge work GDPval-AA v2economically valuable knowledge work (v2, re-based Elo) | 1600 | — |
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. | 1482Best | 1467 |
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. | 1568Best | 1526 |
| Overview | ||
| Company | Anthropic | OpenAI |
| Release date | May 28 2026 | Jun 26 2026 |
| Access | Proprietary | Proprietary |
Which is better: Claude Opus 4.8 or GPT-5.6 Terra?
Claude Opus 4.8 leads GPT-5.6 Terra on 7 of the 9 benchmarks they both report. Claude Opus 4.8 shipped 29 days before GPT-5.6 Terra, so benchmark comparisons should account for the intervening progress.
Published specifications for these two models are limited — see each model page for the latest details.
On BullshitBench v2, Claude Opus 4.8 leads at 95% vs GPT-5.6 Terra at 53%. On Gray Swan IPI · k = 1, Claude Opus 4.8 leads at 0.5% vs GPT-5.6 Terra at 5.4%. On Gray Swan IPI · k = 10, Claude Opus 4.8 leads at 4.1% vs GPT-5.6 Terra at 26%. On Gray Swan IPI · k = 15, Claude Opus 4.8 leads at 5.5% vs GPT-5.6 Terra at 30.4%. On CursorBench v3.2, GPT-5.6 Terra leads at 64.9% vs Claude Opus 4.8 at 62.3%. On Frontier-Bench v0.1, Claude Opus 4.8 leads at 21.1% vs GPT-5.6 Terra at 20.8%. On Terminal-Bench 2.1, GPT-5.6 Terra leads at 84.3% vs Claude Opus 4.8 at 74.6%. On Arena Elo (Text), Claude Opus 4.8 leads at 1482 vs GPT-5.6 Terra at 1467. On Arena Elo (Code), Claude Opus 4.8 leads at 1568 vs GPT-5.6 Terra at 1526.
Frequently asked questions
Claude Opus 4.8 was released by Anthropic on May 28 2026.
GPT-5.6 Terra was released by OpenAI on Jun 26 2026.
GPT-5.6 Terra leads on CursorBench v3.2 — Claude Opus 4.8 62.3% vs GPT-5.6 Terra 64.9%.
Other comparisons
Claude Opus 4.8vsGemini 3.6 FlashGPT-5.6 TerravsGemini 3.6 FlashClaude Opus 4.8vsMuse GlimmerGPT-5.6 TerravsMuse GlimmerClaude Opus 4.8vsGrok 4.5GPT-5.6 TerravsGrok 4.5Claude Opus 4.8vsDeepSeek-V4-Flash-0731GPT-5.6 TerravsDeepSeek-V4-Flash-0731Claude Opus 4.8vsMistral Medium 3.5GPT-5.6 TerravsMistral Medium 3.5Claude Opus 4.8vsKimi K3GPT-5.6 TerravsKimi K3