Reference / 03
Compare models
13 frontier models, side by side: prices per 1M tokens, context, six arena boards, top benchmarks and value score. Pick 2–4 models — the URL is the comparison.
Models · pick 2–4 · 2/4 selected
| Attribute | Claude Fable 5 | Claude Opus 5 |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Open/Closed | Closed | Closed |
| PricingPrice input $/1M | $10.00 | $5.00 |
| Price output $/1M | $50.00 | $25.00 |
| Price cached $/1M | $1.00 | $0.500 |
| Context | 1M | 1M |
| ArenaArena Overall | 1506 | 1493 |
| Arena WebDev | 1627 | 1692 |
| Arena Coding | 1554 | 1531 |
| Arena Hard Prompts | 1532 | 1520 |
| Arena Math | 1527 | 1553 |
| Arena Vision | 1315 | 1297 |
| BenchmarksSWE-bench Pro | 80.0 | — |
| Terminal-Bench | 84.6 | 86.7 |
| GPQA | 92.6 | — |
| HLE | 53.3 | — |
| ValueValue * | 75 | 149 |
| Est. monthly · 10,000 req/day | $10,500.00/mo | $5,250.00/mo |
| Full page → | Full page → |
— not measured / not published · Value * = arena score per $1 of blended price (3:1 input/output mix) · est. monthly = 10,000 req/day × 1,000 in / 500 out tokens × 30 days · estimate with your traffic →