Claude Haiku 4.5 vs Claude Haiku 5.5
| Spec | Claude Haiku 4.5 | Claude Haiku 5.5 |
|---|---|---|
| Respan Index | 19.3 (#132) | 56.8 (#24) |
| Blended price per 1M | $2 | $0.2 |
| Context window | 200K | 1M |
| Input / output per 1M | $1 / $5 | $0.1 / $0.5 |
| Max output | 64K | 128K |
| Input | text, vision | text, vision |
| Released | 2025-10-15 | 2026-10-07 |
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 Haiku 4.5 | Claude Haiku 5.5 |
|---|---|---|
| Coding | 5.1 Claude Haiku 4.5 | 53.8 Claude Haiku 5.5 |
| Agents | 15.7 Claude Haiku 4.5 | 66.0 Claude Haiku 5.5 |
| Reasoning | 21.5 Claude Haiku 4.5 | 64.4 Claude Haiku 5.5 |
| Knowledge & language | 13.2 Claude Haiku 4.5 | 23.7 Claude Haiku 5.5 |
| Human preference | 49.1 Claude Haiku 4.5 | 55.8 Claude Haiku 5.5 |
Benchmarks both report
Claude Haiku 5.5 leads on 13 of the 13 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Claude Haiku 4.5 | Claude Haiku 5.5 |
|---|---|---|
| LMArena Elo | 1414 | 1421 |
| LMArena WebDev (LMArena) | 1330 | 1596 |
| Chartography (no tools) | 6.4% | 46.4% |
| Humanity's Last Exam (no tools) | 10.2% | 45.9% |
| Humanity's Last Exam (with tools) | 18.7% | 57.4% |
| OSWorld 2.1 (offline subset) | 15.7% | 72.4% |
| Terminal-Bench 4.0 | 0% | 39.2% |
| Chess Puzzles (Epoch AI) | 8% | 55% |
| GPQA Diamond | 71.2% | 89.6% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 66.7% | 98.9% |
| SAGE (Vals AI) | 31.8% | 45.8% |
| SimpleQA Verified | 13.2% | 23.8% |
| Vending-Bench 2 (Andon Labs) | 458.89 | 3602.46 |
FAQ
Is Claude Haiku 4.5 better than Claude Haiku 5.5?
Claude Haiku 5.5 scores 56.8 on the Respan Index (#24), ahead of Claude Haiku 4.5 at 19.3 (#132). Claude Haiku 5.5 leads on 13 of the 13 benchmarks both report.
Which is cheaper, Claude Haiku 4.5 or Claude Haiku 5.5?
Claude Haiku 4.5 lists at $1 input and $5 output per 1M tokens; Claude Haiku 5.5 at $0.1 and $0.5. Claude Haiku 5.5 is 10x cheaper per token at list price.
Which has the larger context window?
Claude Haiku 4.5 accepts 200K tokens and Claude Haiku 5.5 1M tokens.