# Claude Haiku 5.5Extra high

> Source: Benchize (https://benchize.com) — https://benchize.com/models/claude-haiku-5-5-xhigh. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/claude-haiku-5-5-xhigh.

Anthropic

Canonical URL: https://benchize.com/models/claude-haiku-5-5-xhigh
Last updated: Thu, 22 Oct 2026 05:40:11 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Anthropic

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Source |  |
| --- | --- | --- | --- | --- | --- | --- |
| [WeirdML v3](https://benchize.com/benchmarks/weirdml%3Av3.md) | 0.12source score | Range 0.1–0.13 | 12 / 25 | — | [weirdml · b0120fe](https://htihle.github.io/assets/data/weirdml_v3.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.11695232727272727 \| source_score \| reported \| Higher is better \| \| `mean_final_best` \| 0.2413118818181818 \| source_score \| reported \| Higher is better \| ### Configuration budget {"kind":"equivalent_tokens","limit":50000000,"start":500000} harness claude_code Reasoning effort xhigh sourceConfiguration haiku-5.5 thinkingEnabled true ### Reported statistics Uncertainty Reported interval: 0.1–0.13 interval [0.10125542544335893,0.13271330598123293] intervalMethod {"assumptions":"Independent fresh runs; approximately normal scores; shared variance prior. Pool fitted within analysis mode, including incomplete models; pool estimation uncertainty is not propagated.","confidence":0.95,"draws":10000,"fixed_tasks":true,"method":"partially_pooled_variance","pool":{"degrees_of_freedom":1533,"fallback":false,"prior_df":4,"variance":0.0026383113392869583},"seed":20260915,"unit":"run"} runCount 75 taskCount 11 ### Source [Original source](https://htihle.github.io/assets/data/weirdml_v3.json) Source weirdml Original model label Claude Haiku 5.5 (xhigh) Source version b0120fe Observed 2026-10-22T04:50:05.075Z Evidence SHA-256 `82c3a8de423b23ca0af9761429299a2f21edb0b819e947fdc863f8b5ea4a49b8` Source locator `/rows/11` |  |  |  |  |  |  |
| [FrontierCode 1.1 extended](https://benchize.com/benchmarks/frontiercode%3Av1_1%3Aextended.md) | 58.12% | — | 46 / 125 | $0.49 | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.5812 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.6413 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 0.4876 \| USD \| per_rollout \| Lower is better \| ### Configuration harness claude-code Reasoning effort xhigh sourceVersion v1_1 subset extended ### Reported statistics flaggedRate 0 sampleSize 150 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label Claude Haiku 5.5 Observed 2026-10-22T05:40:03.220Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1_1/data/Claude%20Haiku%205.5/xhigh/extended` |  |  |  |  |  |  |
| [OTIS Mock AIME 2024–2025](https://benchize.com/benchmarks/epoch-ai%3Aotis_mock_aime_2024_2025.md) | 98.89% | SE ±0.54 pp | 15 / 124 | — | [epoch ai · OTIS Mock AIME 2024-2025](https://epoch.ai/data/benchmark_data.zip#file=otis_mock_aime_2024_2025.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.9888888888888889 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.9888888888888889 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion OTIS Mock AIME 2024-2025 reasoningBudget Not reported Reasoning effort xhigh sourceModelVersion claude-haiku-5-5_xhigh ### Reported statistics Uncertainty Standard error: ±0.54 pp standardError 0.005362818328685681 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=otis_mock_aime_2024_2025.csv) Source epoch ai Original model label claude-haiku-5-5 Source version OTIS Mock AIME 2024-2025 Evaluated 2026-10-07T21:16:36.000Z Observed 2026-10-22T03:00:05.281Z Evidence SHA-256 `b7c9da8c0587c82f1e0d814ceb496c24308380b997a16203b6e5d39eb07f4f50` Source locator `/rows/6` |  |  |  |  |  |  |
| [TapTap Maker public](https://benchize.com/benchmarks/taptap-maker%3Apublic.md) | 75.83% | Range 75.83–75.83% | 17 / 26 | $0.32 | [taptap maker](https://maker.taptap.cn/leaderboard/data/editions/2026-10-18-v6-voxel.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `l2_pass_rate` \| 0.7583 \| score_fraction \| reported \| Higher is better \| \| `cost_per_case` \| 0.3151 \| USD \| per_case \| Lower is better \| ### Configuration datasetVersion 0.2.3 edition 2026-10-18-v6-voxel harness claude Reasoning effort xhigh reasoningMode adaptive ### Reported statistics Uncertainty Reported interval: 75.83–75.83% interval [0.7583,0.7583] intervalMethod bootstrap-case rank 20 rankInterval [17,21] ### Source [Original source](https://maker.taptap.cn/leaderboard/data/editions/2026-10-18-v6-voxel.json) Source taptap maker Original model label claude-haiku-5-5 Observed 2026-10-21T16:33:19.620Z Evidence SHA-256 `ad2d7f1adeae0548a4edc0b5e75d11449b443449ff76533d04bc1fb1b6d7035e` Source locator `/public/19` |  |  |  |  |  |  |
| [FrontierCode 1.1 main](https://benchize.com/benchmarks/frontiercode%3Av1_1%3Amain.md) | 45.83% | — | 44 / 125 | $0.63 | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.4583 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.508 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 0.6297 \| USD \| per_rollout \| Lower is better \| ### Configuration harness claude-code Reasoning effort xhigh sourceVersion v1_1 subset main ### Reported statistics flaggedRate 0 sampleSize 100 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label Claude Haiku 5.5 Observed 2026-10-22T05:40:03.220Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1_1/data/Claude%20Haiku%205.5/xhigh/main` |  |  |  |  |  |  |
| [TapTap Maker heldout](https://benchize.com/benchmarks/taptap-maker%3Aheldout.md) | 58.86% | Range 58.86–58.86% | 15 / 26 | — | [taptap maker](https://maker.taptap.cn/leaderboard/data/editions/2026-10-18-v6-voxel.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `l2_pass_rate` \| 0.5886 \| score_fraction \| reported \| Higher is better \| ### Configuration datasetVersion 0.2.3 edition 2026-10-18-v6-voxel harness claude Reasoning effort xhigh reasoningMode Not reported ### Reported statistics Uncertainty Reported interval: 58.86–58.86% interval [0.5886,0.5886] intervalMethod bootstrap-case rank 17 rankInterval [16,19] ### Source [Original source](https://maker.taptap.cn/leaderboard/data/editions/2026-10-18-v6-voxel.json) Source taptap maker Original model label claude-haiku-5-5 Observed 2026-10-21T16:33:19.620Z Evidence SHA-256 `ad2d7f1adeae0548a4edc0b5e75d11449b443449ff76533d04bc1fb1b6d7035e` Source locator `/heldout/16` |  |  |  |  |  |  |
| [ARC-AGI-1](https://benchize.com/benchmarks/arc-prize%3Av1_Semi_Private.md) | 87.5% | — | 87 / 245 | 0.03usd per task | [arc prize · v1](https://arcprize.org/media/data/leaderboard/v1.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.875 \| score_fraction \| v1_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.03012903599999999 \| usd_per_task \| v1_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v1_Semi_Private harness Not reported modelGroup anthropic-claude-haiku-5-5 modelType CoT Reasoning effort xhigh sourceModelVersion anthropic-claude-haiku-5-5-xhigh thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v1.json) Source arc prize Original model label Claude Haiku 5.5 (XHigh) Source version v1 Observed 2026-10-22T05:20:04.066Z Evidence SHA-256 `2bf68d22dafba37047d07ad833ebc0e8199c9aa2f9eeaed50c75c1b704462836` Source locator `/evaluations/271` |  |  |  |  |  |  |
| [ToneBench Scriptwriting](https://benchize.com/benchmarks/tonebench%3Ascripts.md) | 1,916.2 | Range 1,879–1,954.7 | 44 / 126 | — | [tonebench](https://benchmark.towardsai.com/data/leaderboard.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1916.2 \| rating \| reported \| Higher is better \| \| `overall` \| 80.03 \| percent \| reported \| Higher is better \| \| `anti_slop_numbers` \| 75 \| source_score \| rubric \| Higher is better \| \| `continuity_emotion` \| 78.05 \| source_score \| rubric \| Higher is better \| \| `cue_quality` \| 79.04 \| source_score \| rubric \| Higher is better \| \| `hook_strength` \| 77.49 \| source_score \| rubric \| Higher is better \| \| `length_adherence` \| 95.49 \| source_score \| rubric \| Higher is better \| \| `slop` \| 85.96 \| source_score \| rubric \| Higher is better \| \| `slop_score` \| 96.92 \| source_score \| rubric \| Higher is better \| \| `substance_accuracy` \| 84.47 \| source_score \| rubric \| Higher is better \| \| `tone_voice` \| 75.71 \| source_score \| rubric \| Higher is better \| \| `writing_quality` \| 77.41 \| source_score \| rubric \| Higher is better \| \| `youtube_best_practices` \| 76.16 \| source_score \| rubric \| Higher is better \| ### Configuration provider claude-code Reasoning effort xhigh scoreScope 10_task_current sourceConfiguration claude-haiku-5-5-xhigh thinkingEnabled true ### Reported statistics Uncertainty Reported interval: 1,879–1,954.7 interval [1879,1954.7] rank 44 runCount 5 ### Source [Original source](https://benchmark.towardsai.com/data/leaderboard.json) Source tonebench Original model label Claude Haiku 5.5 (xhigh) Evaluated 2026-10-08 Observed 2026-10-22T04:40:14.102Z Evidence SHA-256 `4264338337f418d3ded7ace250cad64fe12cacbd4b79a312bfaf0ae9268a01c3` Source locator `/rows/43` |  |  |  |  |  |  |
| [ARC-AGI-2](https://benchize.com/benchmarks/arc-prize%3Av2_Semi_Private.md) | 61.25% | — | 75 / 243 | 0.08usd per task | [arc prize · v2](https://arcprize.org/media/data/leaderboard/v2.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.6125 \| score_fraction \| v2_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.08498917333333329 \| usd_per_task \| v2_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v2_Semi_Private harness Not reported modelGroup anthropic-claude-haiku-5-5 modelType CoT Reasoning effort xhigh sourceModelVersion anthropic-claude-haiku-5-5-xhigh thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v2.json) Source arc prize Original model label Claude Haiku 5.5 (XHigh) Source version v2 Observed 2026-10-22T05:20:04.306Z Evidence SHA-256 `4351eb696e78113b0459e9bb5f0258318b905efef34f473dcea51ce641b0d94d` Source locator `/evaluations/272` |  |  |  |  |  |  |

---
Cite as: Claude Haiku 5.5Extra high — Benchize, https://benchize.com/models/claude-haiku-5-5-xhigh (retrieved 2026-10-22)