Claude Opus 5 vs Claude Opus 5.5
| Spec | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| Respan Index | 67.0 (#16) | 76.0 (#4) |
| Blended price per 1M | $10 | $8 |
| Context window | 1M | 1M |
| Input / output per 1M | $5 / $25 | $4 / $20 |
| Max output | 128K | 128K |
| Input | text, vision | text, vision |
| Released | 2026-07-24 | 2026-09-22 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| Coding | 64.0 Claude Opus 5 | 75.9 Claude Opus 5.5 |
| Agents | 70.4 Claude Opus 5 | 78.7 Claude Opus 5.5 |
| Reasoning | 74.0 Claude Opus 5 | 84.1 Claude Opus 5.5 |
| Knowledge & language | 61.1 Claude Opus 5 | 70.4 Claude Opus 5.5 |
| Human preference | 58.5 Claude Opus 5 | 61.6 Claude Opus 5.5 |
Benchmarks both report
Claude Opus 5.5 leads on 24 of the 28 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| LMArena Elo | 1490 | 1506 |
| LMArena Vision (LMArena) | 1285 | 1293 |
| LMArena WebDev (LMArena) | 1692 | 1818 |
| LMArena Agent (LMArena) | 0.0626 | 0.1863 |
| MirrorCode (Epoch AI) | 87.8% | 95.6% |
| Chess Puzzles (Epoch AI) | 43% | 67% |
| APEX-Agents (Mercor) | 65.8% | 73.5% |
| EBR-bench (Epoch AI) | 45.7% | 71.4% |
| FrontierMath Tier 4 v2 (Epoch AI) | 73.2% | 95% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 85.6% | 91.2% |
| FrontierSWE V2 (Proximal Labs) | 52% | 62.3% |
| Furniture Assembly (Epoch AI) | 60.8% | 83.3% |
| GPQA Diamond | 93.9% | 90.6% |
| LiveBench Agentic Coding (LiveBench) | 65.2% | 71.7% |
| LiveBench Coding (LiveBench) | 81.4% | 89.3% |
| LiveBench Data Analysis (LiveBench) | 74.6% | 80.3% |
| LiveBench Instruction Following (LiveBench) | 63.8% | 67% |
| LiveBench Language (LiveBench) | 88.7% | 86.3% |
| LiveBench Mathematics (LiveBench) | 95.7% | 97.1% |
| LiveBench Reasoning (LiveBench) | 91.2% | 92.2% |
| Mystery Game Puzzles (Epoch AI) | 59% | 71% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 98.9% | 100% |
| SAGE (Vals AI) | 49.4% | 45.8% |
| SimpleQA Verified | 59.9% | 72.2% |
| Terminal-Bench 4.0 (Vals AI) | 53.5% | 65.2% |
| Vending-Bench 2 (Andon Labs) | 11181.87 | 9235.25 |
| Terminal-Bench Science 0.1 | 29% | 58.7% |
| SimpleBench (SimpleBench) | 80.6% | 88.4% |
FAQ
Is Claude Opus 5 better than Claude Opus 5.5?
Claude Opus 5.5 scores 76.0 on the Respan Index (#4), ahead of Claude Opus 5 at 67.0 (#16). Claude Opus 5.5 leads on 24 of the 28 benchmarks both report.
Which is cheaper, Claude Opus 5 or Claude Opus 5.5?
Claude Opus 5 lists at $5 input and $25 output per 1M tokens; Claude Opus 5.5 at $4 and $20. Claude Opus 5.5 is 1.3x cheaper per token at list price.
Which has the larger context window?
Both accept 1M tokens of context.