Compare AI models side-by-side
Pick any two models — or up to four — and compare their price per 1M tokens, context window, capabilities and community Arena intelligence score at a glance. The best value in each row is highlighted, and every comparison is a shareable link.
| Claude Opus 5 | Claude Opus 4.6 | |
|---|---|---|
| Intelligence | ||
| Arena · Overall(LMArena Elo — higher is better) | 1504✓ | 1497 |
| Arena rank (overall) | #1✓ | #2 |
| Arena · Coding | 1663✓ | 1536 |
| Arena · Math & reasoning | — | 1425✓ |
| Arena · Agentic / tool use | — | 1435✓ |
| Score evidence | Measured | Measured |
| Pricing (USD / 1M tokens) | ||
| Input | $5.00 | $5.00 |
| Output | $25.00 | $25.00 |
| Cached input (read) | $0.5000 | $0.5000 |
| Cache write | $6.25 | $6.25 |
| Blended(3:1 input:output blend) | $10.00 | $10.00 |
| Specs & capabilities | ||
| Provider | Anthropic | Anthropic |
| Context window | 1M (1,000,000) | 1M (1,000,000) |
| Max output | 128,000✓ | 64,000 |
| Modalities | text | text, image |
| Vision | No | Yes |
| Tool use | Yes | Yes |
| Reasoning | No | Yes |
| Prompt caching | No | Yes |
| Batch pricing | No | Yes |
| Tokenizer | approx | claude |
| Release date | — | — |
| Status | preview | ga |
Intelligence scores are community Arena ratings from LMArena / arena.ai, used under CC BY 4.0. Snapshot last refreshed 27 August 2026. Scores are a relative signal, not an absolute measure of capability. A Measured score was voted on directly; an Estimated score is inherited from an identical base model (a regional/creator-prefixed hosting duplicate). See our methodology.
Popular comparisons
Jump straight to a detailed head-to-head page for these popular pairings.
FAQ
How do I compare two AI models?
Search for the first model, then the second, and this tool shows their price per 1M tokens, context window, capabilities and Arena intelligence score side-by-side. Add up to four models to compare at once.
Which model is cheaper?
The cheapest cell in each pricing row (input, output, cached read, cache write and the blended 3:1 estimate) is highlighted, so you can see at a glance which model costs less for reads versus writes.
Is the comparison shareable?
Yes. Your selection is stored in the URL (?models=a,b), so copying the link shares the exact comparison you're looking at.
Estimates for planning only. Verify current pricing with each provider.