Claude Opus 5.5vsGPT-6 Sol
Claude Opus 5.5 | GPT-6 Sol | |
|---|---|---|
| Specifications | ||
Context windowHow much text the model can hold in mind at once — your question, any documents you attach, the conversation so far, and its own reply. Go past it and the earliest part falls out of view. | 1M | 1.05M |
| API pricingUSD per 1M tokens · lower wins · base tier | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $4.00 | $2.00 |
Output priceWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one. | $20.00 | $10.00 |
Cached input priceA 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. | $0.20 | $0.20 |
| Benchmarks | ||
OSWorld 2.0Agentic computer use — Can 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. | 81.8% | 60.5% |
AutomationBenchBusiness workflows — Tests 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. | 40% | 33.2% |
| BenchmarksPublished by one model only | ||
DeepSWE 1.1Agentic 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. (Version 1.1 of the test.) Higher is better. | — | 68.8% |
FrontierCode v1.1 (Main) · main splitAgentic coding — 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. | 54.4% | — |
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. | 66.4% | — |
Terminal-Bench-Science 0.1Agentic scientific computing — The same command-line setup as Terminal-Bench, pointed at scientific work: the AI has to drive research tooling and computational workflows through to a result, rather than administer a machine. Version 0.1 is the first release of the task set, and scores run lower than on the general board. Higher is better. | 58.7% | — |
Humanity's Last Exam · with toolsMultidisciplinary reasoning — Humanity'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. | 67.7% | — |
Agent's Last Exam · pass@1Agentic computer use — A hard set of desktop and operating-system tasks an AI agent has to finish by looking at the screen and working the machine itself. The score is the share it passes outright — partial credit does not count. Higher is better. | — | 56.4% |
GDPval-AA v2.1Knowledge work — economically valuable knowledge work (v2.1, Crowd-BT Elo fit) | 1846 | — |
Chartography · with toolsChart tasks — A chart-centred test run with tools available to the AI, reported separately from the chart-reading benchmarks above it. Higher is better. | 89% | — |
| Overview | ||
| Company | Anthropic | OpenAI |
| Release date | Sep 22 2026 | Sep 22 2026 |
| Access | Proprietary | Proprietary |
Other comparisons
Claude Opus 5.5vsGemini 3.8 FlashGPT-6 SolvsGemini 3.8 FlashClaude Opus 5.5vsMuse Spark 1.3GPT-6 SolvsMuse Spark 1.3Claude Opus 5.5vsGrok 4.7GPT-6 SolvsGrok 4.7Claude Opus 5.5vsDeepSeek-V4.1-FlashGPT-6 SolvsDeepSeek-V4.1-FlashClaude Opus 5.5vsMistral Medium 3.5GPT-6 SolvsMistral Medium 3.5Claude Opus 5.5vsKimi K3GPT-6 SolvsKimi K3
Frequently asked questions
Claude Opus 5.5 leads GPT-6 Sol on 2 of the 2 benchmarks they both report (OSWorld 2.0, AutomationBench). GPT-6 Sol is cheaper on both input and output: $2.00 vs $4.00 per million input tokens, and $10.00 vs $20.00 per million output tokens. Figures are base-tier rates. Both models were released on the same day — Sep 22 2026.
Context windows are 1M (Claude Opus 5.5) vs 1.05M (GPT-6 Sol).
On OSWorld 2.0, Claude Opus 5.5 leads at 81.8% vs GPT-6 Sol at 60.5%. On AutomationBench, Claude Opus 5.5 leads at 40% vs GPT-6 Sol at 33.2%.