Claude Haiku 5.5 Max
| Released | 2026-10-07 |
|---|---|
| Price per 1M | $0.10 in · $0.50 out |
| Overall score | 60.1 · rank 92 |
| Context window | 1,000,000 tokens |
| Provider | Anthropic |
| Weights | Proprietary |
| Configuration | Max |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 60.1 | 92 | 32 |
| Overall: Coding | 61.9 | 65 | 27 |
| Overall: Reasoning | 58.9 | 135 | 13 |
| Overall: Agentic work | 61.2 | 74 | 26 |
| Overall: Frontend | 62.8 | 40 | 10 |
| Coding | 62.1 | 54 | 9 |
| Agents | 60.7 | 56 | 9 |
| Reasoning | 61.6 | 97 | 8 |
| Math | 61.1 | 40 | 3 |
| Knowledge | 49.3 | 133 | 2 |
| Frontend | 63.1 | 16 | 1 |
| Value vs price | +14.3 | 15 |
Evidence
Coding 62.1 ±1.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CursorBench | 48.4% | Epoch AI | 2026-11-11 |
| FrontierSWE | 43.8% | Epoch AI | 2026-11-11 |
| FrontierCode | 46.4% | Epoch AI | 2026-11-11 |
| Code Migration v1.1 | 52.4% | Vals AI | 2026-11-09 |
| MirrorCode | 78.9% | Epoch AI | 2026-10-21 |
| IOI | 47.3% | Vals AI | 2026-10-29 |
| SWE-bench Multimodal | 44.0% | Vals AI | 2026-11-09 |
| SciCode | 55.0% | Epoch AI | 2026-11-11 |
| ProgramBench | 2.5% | Vals AI | 2026-11-09 |
Agents 60.7 ±0.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 4.0 (Vals run) | 36.4% | Vals AI | 2026-11-09 |
| SRE Bench | 5.7% | Vals AI | 2026-11-05 |
| Terminal-Bench-Science | 10.0% | Vals AI | 2026-11-09 |
| APEX-Agents | 56.1% | Epoch AI | 2026-11-11 |
| EMB v1.1 (Excel financial modeling) | 65.5% | Vals AI | 2026-11-09 |
| Tax Agent Bench | 31.9% | Vals AI | 2026-11-09 |
| Finance Agent v2.1 | 65.4% | Vals AI | 2026-11-09 |
| HLAB-Vals 1.1 (legal agent) | 4.2% | Vals AI | 2026-11-09 |
| CyberBench v1.1 | 75.5% | Vals AI | 2026-11-05 |
Reasoning 61.6 ±1.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MysteryMechanism | 35.1% | Vals AI | 2026-11-09 |
| GPQA Diamond | 89.7% | Epoch AI | 2026-10-09 |
| CritPt | 18.9% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 89.0% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 89.0% | Epoch AI | 2026-11-11 |
| BioMysteryBench | 70.4% | Vals AI | 2026-11-09 |
| ARC-AGI-3 | 2.8% | ARC Prize | 2026-11-11 |
| Chess Puzzles | 55.0% | Epoch AI | 2026-10-13 |
| Mystery Game Puzzles | 30.0% | Epoch AI | 2026-10-07 |
Math 61.1 ±1.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 75.1% | Epoch AI | 2026-10-07 |
| FrontierMath Tier 4 | 46.3% | Epoch AI | 2026-10-07 |
| ProofBench v1.1 | 89.0% | Vals AI | 2026-11-09 |
| ProofBench | 89.0% | Epoch AI | 2026-11-11 |
Knowledge 49.3 ±2.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 23.8% | Epoch AI | 2026-10-09 |
| Legal Research Bench | 43.8% | Vals AI | 2026-11-09 |
Frontend 63.1 ±0.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Vibe Code Bench v1.1 | 90.4% | Vals AI | 2026-11-09 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-3 | 0.5% | ARC Prize | Lower duplicate run on the same benchmark |
Names matched to this model
claude-haiku-5-5-max