Head to head
Claude Opus 4.6 vs Claude Opus 4.5
Claude Opus 4.6 from Anthropic against Claude Opus 4.5 from Anthropic — specification, price and every benchmark both makers have published, in one table.
At a glance
Benchmarks
Evenly split
Across the 2 benchmarks both models report, they are level — Claude Opus 4.6 takes 1 and Claude Opus 4.5 takes 1. Which one is better depends entirely on which test resembles your work.
Price
Same price
Both cost $10.00 per million tokens on a 3:1 input-to-output mix.
What actually differs
- Claude Opus 4.6 takes 1M tokens of context against 200K — 5.0× more room for long documents or a large codebase.
Scorecard
Which is better at what
Maths, coding, reasoning and the rest — one line each, averaged over the benchmarks both models actually report.
| Category | Claude Opus 4.6 | Claude Opus 4.5 | Better at this |
|---|---|---|---|
| ReasoningGPQA Diamond | 91.3% | 87.0% | Claude Opus 4.6+4.3 |
| MathsOnly one model reports this | — | — | Not comparable |
| CodingSWE-bench Verified | 80.8% | 80.9% | Claude Opus 4.5+0.1 |
| KnowledgeNeither model reports this | — | — | Not comparable |
| MultimodalOnly one model reports this | — | — | Not comparable |
| Instruction followingNeither model reports this | — | — | Not comparable |
| Human preferenceNeither model reports this | — | — | Not comparable |
| Categories wonOut of 2 comparable | 1 | 1 | Split — Claude Opus 4.6 and Claude Opus 4.5 |
Each category averages only the benchmarks every model here reports, so no one is credited for a test the other did not run. A category with no shared test is marked Not comparable rather than guessed at.
Side by side
Claude Opus 4.6 and Claude Opus 4.5, row by row
| Attribute | Claude Opus 4.6Anthropic | Claude Opus 4.5Anthropic |
|---|---|---|
| Specification | ||
| MakerWho built it | Anthropic | Anthropic |
| Released | 2026-02 | 2025-11 |
| Context windowHow much can go in at once | 1,000,000 tokens | 200,000 tokens |
| Max output | 128,000 tokens | 64,000 tokens |
| Input | Text, Image | Text, Image |
| ReasoningSpends extra tokens thinking before it answers | Yes | Yes |
| Tool calling | Yes | Yes |
| Knowledge cutoff | 2025-05 | 2025-03 |
| Licence | Proprietary | Proprietary |
| Open weightsCan you download and run it yourself | No | No |
| Price | ||
| Input priceUSD per million tokens in | $5.00 | $5.00 |
| Output priceUSD per million tokens out | $25.00 | $25.00 |
| Cached input | $0.50 | $0.50 |
| Blended 3:1A 3-in-to-1-out million-token mix — a fairer single number than input price alone | $10.00 | $10.00 |
| Published benchmarks | ||
| GPQA Diamond198 graduate-level physics, chemistry and biology questions written to be Google-proof. PhD holders in the matching field score about 65%. | ||
| AIME 2025The American Invitational Mathematics Examination — 15 problems, integer answers, no partial credit. A standard test of multi-step maths reasoning. | Not reported | |
| SWE-bench Verified500 human-validated GitHub issues from real Python repositories. The model must produce a patch that makes the project's own tests pass. | ||
| Terminal-Bench 2.0The 2.0 task set of the terminal agent benchmark — the version most launches reported from late 2025 to spring 2026. Not comparable with 2.1 or 4.0. | Not reported | |
| MMMUCollege-level questions that require reading charts, diagrams, tables and photographs alongside the text. | Not reported | |
| Links | Full page | Full page |
| Row verified | 2026-10 | 2026-08 |
Where these numbers come from
Every score on this page is a published figure, taken from the model's own card, system card, technical report or release post, or from a public leaderboard. CorX Labs did not run these evaluations. Most are self-reported by the lab that built the model, which means they were produced under that lab's own choice of prompt, scaffold and number of attempts — so treat them as a starting point for a shortlist, not as a settled ranking.
A score someone other than the model's maker measured is marked Independent and names its measurer. Those are the stronger numbers on this page — an outside harness has no reason to flatter anyone — and there are not many of them.
Where a figure has not been published, the cell reads Not reported rather than an estimate. Nothing here is inferred, interpolated or guessed. Each model records the month its row was last checked. Full method and caveats.
Questions
Claude Opus 4.6 or Claude Opus 4.5?
Is Claude Opus 4.6 better than Claude Opus 4.5?
Across the 2 benchmarks both models report, they are level — Claude Opus 4.6 takes 1 and Claude Opus 4.5 takes 1. Which one is better depends entirely on which test resembles your work.
Which is cheaper, Claude Opus 4.6 or Claude Opus 4.5?
Both cost $10.00 per million tokens on a 3:1 input-to-output mix.
What is the difference between Claude Opus 4.6 and Claude Opus 4.5?
Claude Opus 4.6 takes 1M tokens of context against 200K — 5.0× more room for long documents or a large codebase.
