Claude Haiku 5.5 (max): benchmark scores, price and ranking
Claude Haiku 5.5 (max) from Anthropic ranks #82 of 743 LLMs on the LLMs Tiger aggregate benchmark score (52.1), based on 7 benchmark results captured up to 2026-10-13.
- Organisation
- Anthropic
- Type
- Fast / low-cost (flash) model
- Open weights
- No
- Released
- 2026-10-07 (source: epoch)
- Cheapest API price
- $0.100 in / $0.500 out per million tokens (via anthropic, observed 2026-10-13T20:47Z)
- Aggregate rank
- #82 of 743 (score 52.1)
Relative to other models, Claude Haiku 5.5 (max) is strongest on GPQA Diamond (#38 of 275) and weakest on SimpleQA Verified (#54 of 69).
Across the 7 benchmarks where at least ten models are scored, Claude Haiku 5.5 (max) is in the top 10% on 0 and in the top half on 6.
Its cheapest listed input price, $0.100 per million tokens, is lower than that of 79% of the 427 models with a listed price.
Claude Haiku 5.5 (max) benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| GPQA Diamond | 89.6% | #38 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| FrontierMath T1-3 | 75.1% | #15 of 68 | Epoch AI Benchmarking Hub | 2026-10-13 |
| FrontierMath T4 | 46.3% | #16 of 50 | Epoch AI Benchmarking Hub | 2026-10-13 |
| EBR-Bench | 43.8% | #8 of 21 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Mystery Games | 30.0% | #23 of 59 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Furniture Assembly | 47.5% | #11 of 28 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SimpleQA Verified | 23.8% | #54 of 69 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Claude Haiku 5.5 (max)
Claude Haiku 5.5 (max) is a smaller, lower-cost model from Anthropic. It was released on 2026-10-07, according to Epoch AI. The cheapest API listing we track charges $0.100 per million input tokens and $0.500 per million output tokens, through anthropic. We hold 7 benchmark results for it across 2 areas, and none is in the top tenth of its board.
Where Claude Haiku 5.5 (max) stands
On our aggregate score, Claude Haiku 5.5 (max) is #82 of 743 ranked models, in the top quarter. Its score is 52.1; the scale is centred on 50, so it sits 2.1 points above the typical model. That beats 89% of the ranked field. Among the 70 ranked Anthropic models it is #18. The next one up is Claude Opus 4.6 (max) at 52.5. The next one down is Claude 4.7 (Max) at 51.8.
Claude Haiku 5.5 (max), benchmark by benchmark
GPQA Diamond puts Claude Haiku 5.5 (max) at 89.6%, #38 of 275. That beats 86% of the models tested, but 6.1 points short of GPT-6 Astra (max) (95.8%). The results either side are Muse Spark (89.8%) and DeepSeek v4 Pro (max) (89.6%). It covers expert-level science questions. The board median is 71.7%, so Claude Haiku 5.5 (max) is 17.9 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On FrontierMath T1-3, Claude Haiku 5.5 (max) reaches 75.1%. It ranks #15 of 68, ahead of 78% of the field, and 18.6 points behind GPT-6 Astra (max) at 93.7%. The results either side are GPT-5.4 (xhigh) (78.6%) and Qwen3.8 Max (xhigh) (74.7%). It covers research-level maths, tiers 1-3. The board median is 55.6%, so Claude Haiku 5.5 (max) is 19.5 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Claude Haiku 5.5 (max) manages 46.3% on FrontierMath T4, good for #16 of 50. GPT-6.1 Sol (max) leads at 100.0%, 53.7 points further up, and 68% of models sit below. The results either side are GPT-5.4 (xhigh) (49.0%) and Muse Spark 1.3 (max) (46.3%). It covers research-level maths, tier 4. The board median is 30.5%, so Claude Haiku 5.5 (max) is 15.9 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
EBR-Bench puts Claude Haiku 5.5 (max) at 43.8%, #8 of 21. That beats 62% of the models tested, but 32.4 points short of GPT-6 Astra (max) (76.2%). The results either side are GPT-5.6 Sol (max) (44.8%) and Claude Fable 5 (max) (39.5%). It covers evidence-based reasoning. The board median is 31.4%, so Claude Haiku 5.5 (max) is 12.4 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On Mystery Games, Claude Haiku 5.5 (max) reaches 30.0%. It ranks #23 of 59, ahead of 61% of the field, and 54.0 points behind GPT-6 Astra (max) at 84.0%. The results either side are Qwen3.7 Max (32.0%) and Grok 4.7 (xhigh) (29.0%). It covers deductive reasoning puzzles. The board median is 23.0%, so Claude Haiku 5.5 (max) is 7.0 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Claude Haiku 5.5 (max) manages 47.5% on Furniture Assembly, good for #11 of 28. Claude Opus 5.5 (max) leads at 83.3%, 35.8 points further up, and 61% of models sit below. The results either side are GPT-5.6 Terra (max) (54.2%) and GPT-5.5 (xhigh) (44.2%). It covers planning a physical assembly. The board median is 41.2%, so Claude Haiku 5.5 (max) is 6.2 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Claude Haiku 5.5 (max) lands below the middle on SimpleQA Verified: 23.8%, #54 of 69, with 77% of the field scoring higher. gemini-4-argon-high leads at 76.0%; the closest result above is Qwen3.5 Plus at 25.4%. It covers short factual questions. The board median is 41.0%, so Claude Haiku 5.5 (max) is 17.2 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: strong, ahead of 77% of the field on average across 3 benchmarks. Its best board here is GPQA Diamond (#38 of 275) and its weakest is FrontierMath T4 (#16 of 50). General knowledge and chat: solid, ahead of 51% of the field on average across 4 benchmarks. Its best board here is EBR-Bench (#8 of 21) and its weakest is SimpleQA Verified (#54 of 69).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Claude Haiku 5.5 (max) costs less than the typical listing. Models that score within 1.5 points or better and cost less: DeepSeek V4 Flash 0731 (max) ($0.016, 53.1), GLM-5.2 (high) ($0.060, 51.6), Qwen3-Coder-30B-A3B-Instruct ($0.070, 52.7). Similar scores cost more on Qwen3.5-122B-A10B ($0.113) and MiMo-V2.6 Flash ($0.140).
Release history
The Anthropic model released before it was Claude Sonnet 5.5 (max) on 2026-09-28, which scores 60.8.
Reading these numbers
The 7 scores on this page come from 1 source: Epoch AI Benchmarking Hub (7). No official leaderboard supplied any of them, so treat them as third-party measurements. The leaderboard also lists 2 other reasoning-effort settings of this model, which redirect here. This page shows the highest. With 7 results the aggregate rank can move a lot when one new score arrives.
Claude Haiku 5.5 (max) is also listed at 2 other reasoning-effort settings; this page shows the highest-effort row, which is the one the leaderboard keeps by default.
Models ranked nearby
- #79 Qwen3.8-2.4T-A95B (52.1)
- #80 gpt-5.2-2025-12-11-high (52.1)
- #81 EXAONE-4.5-33B (52.1)
- #83 NVIDIA-Nemotron-3-Super-120B-A12B-BF16 (52.0)
- #84 K-EXAONE-236B-A23B (52.0)
- #85 Grok 4.5 (high) (52.0)
More from Anthropic
- Claude Opus 5.5 (max) (#1, 64.3)
- Claude Fable 5.1 (max) (#3, 61.9)
- Claude Sonnet 5.5 (max) (#4, 60.8)
- Claude Opus 5 (max) (#8, 60.0)
- Claude Fable 5 (max) (#9, 59.9)
- claude-opus-4-8-xhigh-effort (#14, 56.6)
Compare it on the live leaderboard → · How the aggregate score works