# Claude Haiku 4.5

> Source: Benchize (https://benchize.com) — https://benchize.com/models/claude-haiku-4-5. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/claude-haiku-4-5.

Anthropic

Canonical URL: https://benchize.com/models/claude-haiku-4-5
Last updated: Thu, 22 Oct 2026 06:30:09 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 |  |
| --- | --- | --- | --- | --- | --- | --- |
| [ToneBench Scriptwriting](https://benchize.com/benchmarks/tonebench%3Ascripts.md) | 839.3 | Range 785.4–903.2 | 106 / 126 | — | [tonebench](https://benchmark.towardsai.com/data/leaderboard.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 839.3 \| rating \| reported \| Higher is better \| \| `overall` \| 65.41 \| percent \| reported \| Higher is better \| \| `anti_slop_numbers` \| 58.02 \| source_score \| rubric \| Higher is better \| \| `continuity_emotion` \| 66.61 \| source_score \| rubric \| Higher is better \| \| `cue_quality` \| 51.82 \| source_score \| rubric \| Higher is better \| \| `hook_strength` \| 73.35 \| source_score \| rubric \| Higher is better \| \| `length_adherence` \| 67.81 \| source_score \| rubric \| Higher is better \| \| `slop` \| 72.83 \| source_score \| rubric \| Higher is better \| \| `slop_score` \| 87.64 \| source_score \| rubric \| Higher is better \| \| `substance_accuracy` \| 66.34 \| source_score \| rubric \| Higher is better \| \| `tone_voice` \| 62.16 \| source_score \| rubric \| Higher is better \| \| `writing_quality` \| 64.6 \| source_score \| rubric \| Higher is better \| \| `youtube_best_practices` \| 62.08 \| source_score \| rubric \| Higher is better \| ### Configuration provider claude-code Reasoning effort unknown scoreScope 10_task_current sourceConfiguration claude-haiku-4-5 ### Reported statistics Uncertainty Reported interval: 785.4–903.2 interval [785.4,903.2] rank 110 runCount 5 ### Source [Original source](https://benchmark.towardsai.com/data/leaderboard.json) Source tonebench Original model label Claude Haiku 4.5 (extended thinking) Evaluated 2026-10-08 Observed 2026-10-22T04:40:14.102Z Evidence SHA-256 `4264338337f418d3ded7ace250cad64fe12cacbd4b79a312bfaf0ae9268a01c3` Source locator `/rows/109` |  |  |  |  |  |  |
| [Judgemark v2.1](https://benchize.com/benchmarks/eq-bench%3Ajudgemark-v2.md) | 79.72% | — | 16 / 39 | — | [eq bench · judgemark-v2](https://eqbench.com/judgemark-v2.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `judgemark_score` \| 79.72 \| percent \| overall \| Higher is better \| \| `stability` \| 0.781 \| score_fraction \| overall \| Higher is better \| \| `separability` \| 0.814 \| score_fraction \| overall \| Higher is better \| \| `human_corr` \| 0.747 \| correlation \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol judgemark-v2 sourceModelVersion claude-haiku-4.5 ### Source [Original source](https://eqbench.com/judgemark-v2.js) Source eq bench Original model label claude-haiku-4.5 Source version judgemark-v2 Observed 2026-10-22T05:10:06.632Z Evidence SHA-256 `070cc3fbd6f0333dde2b52bec4329ea3fba7eccbf016bbbbef95e0f8b2197781` Source locator `/leaderboardDataV2/23` |  |  |  |  |  |  |
| [DesignArena 3D design generation](https://benchize.com/benchmarks/designarena%3A3d-design.md) | 1,072elo | — | 75 / 110 | — | [designarena · public](https://www.designarena.ai/api/leaderboard#board=3d-design) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1072 \| elo \| 3d-design \| Higher is better \| ### Configuration arenaType models category 3d inputModality Not reported sourceModelVersion claude-haiku-4-5 variationName public ### Reported statistics averageGenerationTimeMs 29732 battles 6343 losses 3738 standardError Not reported winRate 41.1 wins 2605 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=3d-design) Source designarena Original model label Claude Haiku 4.5 Source version public Observed 2026-10-21T16:37:58.515Z Evidence SHA-256 `bcc0267f7bc7e9ec112306b3a35413cf78bbccbfaa838356232d5d5ed0b83d0f` Source locator `/data/119` |  |  |  |  |  |  |
| [NC Bench](https://benchize.com/benchmarks/nc-bench%3Aoverall.md) | 83.36% | — | 113 / 144 | — | [nc bench](https://www.nc-bench.com/models) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 83.36 \| percent \| reported \| Higher is better \| \| `tooling` \| 96.75 \| percent \| category \| Higher is better \| \| `creative_writing` \| 78.96 \| percent \| category \| Higher is better \| \| `language` \| 91.84 \| percent \| category \| Higher is better \| \| `utility` \| 72.48 \| percent \| category \| Higher is better \| \| `reasoning` \| 67.77 \| percent \| category \| Higher is better \| \| `text_editing` \| 96.81 \| percent \| category \| Higher is better \| \| `rule_following` \| 70.35 \| percent \| category \| Higher is better \| \| `hallucination` \| 91.93 \| percent \| category \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration Claude Haiku 4.5 thinkingEnabled Not reported ### Source [Original source](https://www.nc-bench.com/models) Source nc bench Original model label Claude Haiku 4.5 Observed 2026-10-22T06:10:03.426Z Evidence SHA-256 `ec261022458543624b08a136d01f5a2f0666cd52ffb49c64d6ca983e8543d1b5` Source locator `/rows/146` |  |  |  |  |  |  |
| [ARC-AGI-2](https://benchize.com/benchmarks/arc-prize%3Av2_Semi_Private.md) | 1.25% | — | 213 / 243 | 0.04usd per task | [arc prize · v2](https://arcprize.org/media/data/leaderboard/v2.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.0125 \| score_fraction \| v2_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.0426 \| usd_per_task \| v2_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v2_Semi_Private harness Not reported modelGroup claude-haiku-4-5-20251001 modelType Base LLM sourceModelVersion claude-haiku-4-5-20251001 ### Source [Original source](https://arcprize.org/media/data/leaderboard/v2.json) Source arc prize Original model label Claude Haiku 4.5 Source version v2 Observed 2026-10-22T05:20:04.306Z Evidence SHA-256 `4351eb696e78113b0459e9bb5f0258318b905efef34f473dcea51ce641b0d94d` Source locator `/evaluations/83` |  |  |  |  |  |  |
| [Vending-Bench 2](https://benchize.com/benchmarks/vendingbench%3Av2.md) | — | — | — | $458.89 | [vendingbench](https://andonlabs.com/_app/immutable/chunks/Cjn0_qj9.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `final_balance` \| 458.88599999999985 \| USD \| one_simulated_year \| Higher is better \| \| `geometric_mean_balance` \| 364.15229067740256 \| USD \| positive_final_balances \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration Claude Haiku 4.5 thinkingEnabled Not reported ### Reported statistics confidenceHalfWidth95 538.8890147491016 positiveBalanceRuns 4 runCount 5 standardDeviation 614.7920761769786 standardError 274.9433748719906 ### Source [Original source](https://andonlabs.com/_app/immutable/chunks/Cjn0_qj9.js) Source vendingbench Original model label Claude Haiku 4.5 Observed 2026-10-21T16:36:40.592Z Evidence SHA-256 `1a0b043147050298c405bcb743d73e425d532464202231cd18c27f836fa5f223` Source locator `/rows/6` |  |  |  |  |  |  |
| [DesignArena Overall frontend coding](https://benchize.com/benchmarks/designarena%3Acode.md) | 1,115elo | SE ±2.01 | 87 / 116 | — | [designarena · public](https://www.designarena.ai/api/leaderboard#board=code) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1115 \| elo \| code \| Higher is better \| ### Configuration arenaType models category allcategories inputModality text sourceModelVersion claude-haiku-4-5 variationName public ### Reported statistics Uncertainty Standard error: ±2.01 averageGenerationTimeMs 48104 battles 35541 losses 19589 standardError 2.0116 winRate 44.9 wins 15952 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=code) Source designarena Original model label Claude Haiku 4.5 Source version public Observed 2026-10-21T16:37:57.301Z Evidence SHA-256 `0c897b58670f5e352bd123db999a40c0d1781a443964c502b90774a76a3a7758` Source locator `/data/131` |  |  |  |  |  |  |
| [DesignArena SVG generation](https://benchize.com/benchmarks/designarena%3Asvgs.md) | 1,026elo | — | 69 / 77 | — | [designarena · public](https://www.designarena.ai/api/leaderboard#board=svgs) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1026 \| elo \| svgs \| Higher is better \| ### Configuration arenaType models category svg inputModality Not reported sourceModelVersion claude-haiku-4-5 variationName public ### Reported statistics averageGenerationTimeMs 11562 battles 4714 losses 2872 standardError Not reported winRate 39.1 wins 1842 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=svgs) Source designarena Original model label Claude Haiku 4.5 Source version public Observed 2026-10-21T16:38:03.249Z Evidence SHA-256 `8cbc34a474a66ef5c64f3ffe557df9b55bf7e13368c07c0214223e3ed8f3f21e` Source locator `/data/109` |  |  |  |  |  |  |
| [DesignArena Data visualization generation](https://benchize.com/benchmarks/designarena%3Adata-viz.md) | 1,124elo | — | 81 / 99 | — | [designarena · public](https://www.designarena.ai/api/leaderboard#board=data-viz) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1124 \| elo \| data-viz \| Higher is better \| ### Configuration arenaType models category dataviz inputModality Not reported sourceModelVersion claude-haiku-4-5 variationName public ### Reported statistics averageGenerationTimeMs 39176 battles 1131 losses 615 standardError Not reported winRate 45.6 wins 516 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=data-viz) Source designarena Original model label Claude Haiku 4.5 Source version public Observed 2026-10-21T16:37:58.264Z Evidence SHA-256 `a895f6f1a25fa8eb01f821c2097ddec4f884de7bd22f1b8a22e82e6679b71849` Source locator `/data/118` |  |  |  |  |  |  |
| [DesignArena Website generation](https://benchize.com/benchmarks/designarena%3Awebsite.md) | 1,122elo | — | 90 / 123 | — | [designarena · public](https://www.designarena.ai/api/leaderboard#board=website) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1122 \| elo \| website \| Higher is better \| ### Configuration arenaType models category website inputModality text sourceModelVersion claude-haiku-4-5 variationName public ### Reported statistics averageGenerationTimeMs 52499 battles 23935 losses 13138 standardError Not reported winRate 45.1 wins 10797 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=website) Source designarena Original model label Claude Haiku 4.5 Source version public Observed 2026-10-21T16:37:57.757Z Evidence SHA-256 `51873e97776a3c5ae706b24cc2c60384793dd18398df970e213a8e7c276d8ac0` Source locator `/data/134` |  |  |  |  |  |  |
| [ARC-AGI-1](https://benchize.com/benchmarks/arc-prize%3Av1_Semi_Private.md) | 14.33% | — | 222 / 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` \| 0.1433 \| score_fraction \| v1_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.0263 \| usd_per_task \| v1_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v1_Semi_Private harness Not reported modelGroup claude-haiku-4-5-20251001 modelType Base LLM sourceModelVersion claude-haiku-4-5-20251001 ### Source [Original source](https://arcprize.org/media/data/leaderboard/v1.json) Source arc prize Original model label Claude Haiku 4.5 Source version v1 Observed 2026-10-22T05:20:04.066Z Evidence SHA-256 `2bf68d22dafba37047d07ad833ebc0e8199c9aa2f9eeaed50c75c1b704462836` Source locator `/evaluations/81` |  |  |  |  |  |  |
| [SVGBench Arena](https://benchize.com/benchmarks/svgbench%3Aarena.md) | 1,532 | — | 58 / 191 | — | [svgbench](https://svgbench.ai/api/leaderboard?shape=2) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1532 \| rating \| reported \| Higher is better \| ### Configuration Reasoning effort unknown sourceModelId model_claudehaiku45 ### Reported statistics coverageCount 27 draws 7 judged 226 losses 62 rank 72 ratingDeviation 28 wins 157 ### Source [Original source](https://svgbench.ai/api/leaderboard?shape=2) Source svgbench Original model label Claude Haiku 4.5 Observed 2026-10-22T04:30:04.538Z Evidence SHA-256 `5c0a9c869593820d57d2c05f6d416e1b44a740b4835e66c66e9ad8b8edc59e2c` Source locator `/rows/69` |  |  |  |  |  |  |
| [AutomationBench](https://benchize.com/benchmarks/zapier%3Aautomationbench.md) | 0.46% | — | 107 / 113 | $0.07 | [zapier](https://framerusercontent.com/sites/4WTSl4BNjd1q9QFEFibC6h/lOvLnhItj8S8-1vyjH_s33Jn-WmUXPmiJ5Diajl0Q38.CMutaHYl.mjs) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.46 \| percent \| reported \| Higher is better \| \| `cost` \| 0.07 \| USD \| per_task \| Lower is better \| ### Configuration benchmarkVersion 1.0.6 costFootnote false Fallback policy Not reported Reasoning effort unknown Model configuration Claude Haiku 4.5 thinkingEnabled Not reported ### Reported statistics rank 120 ### Source [Original source](https://framerusercontent.com/sites/4WTSl4BNjd1q9QFEFibC6h/lOvLnhItj8S8-1vyjH_s33Jn-WmUXPmiJ5Diajl0Q38.CMutaHYl.mjs) Source zapier Original model label Claude Haiku 4.5 Observed 2026-10-21T16:34:56.783Z Evidence SHA-256 `43978e38f018789fa59c11de5328abe9e5905bfa9a5a4d601fa3aa4d9c153b0d` Source locator `/rows/119` |  |  |  |  |  |  |
| [Longform Creative Writing](https://benchize.com/benchmarks/eq-bench%3Alongform-writing.md) | 65% | — | 46 / 93 | — | [eq bench · longform-writing](https://eqbench.com/creative_writing_longform.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `overall_score_100` \| 65.0 \| percent \| overall \| Higher is better \| \| `chapter1_avg` \| 15.37 \| rubric_score \| overall \| Higher is better \| \| `chapter2_avg` \| 14.60 \| rubric_score \| overall \| Higher is better \| \| `chapter3_avg` \| 14.41 \| rubric_score \| overall \| Higher is better \| \| `chapter4_avg` \| 14.16 \| rubric_score \| overall \| Higher is better \| \| `chapter5_avg` \| 13.39 \| rubric_score \| overall \| Higher is better \| \| `chapter6_avg` \| 12.54 \| rubric_score \| overall \| Higher is better \| \| `chapter7_avg` \| 11.02 \| rubric_score \| overall \| Higher is better \| \| `chapter8_avg` \| 10.86 \| rubric_score \| overall \| Higher is better \| \| `final_judgement_avg` \| 12.39 \| rubric_score \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol longform-writing sourceModelVersion claude-haiku-4.5 ### Source [Original source](https://eqbench.com/creative_writing_longform.js) Source eq bench Original model label claude-haiku-4.5 Source version longform-writing Observed 2026-10-22T05:10:05.903Z Evidence SHA-256 `3ba283ce357ac7620e16db640a961879c30fe3ddc50a27de9a8ed4290784825c` Source locator `/leaderboardDataLongformV3/20` |  |  |  |  |  |  |
| [MCP Atlas](https://benchize.com/benchmarks/scale%3Amcp_atlas.md) | 40.2source score | CI 37.2–43.2 | 22 / 22 | — | [scale](https://labs.scale.com/leaderboard/mcp_atlas) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 40.2 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration claude-haiku-4-5 sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 37.2–43.2 contaminationWarning Not reported rank 34 reportedConfidenceHalfWidth 3 ### Source [Original source](https://labs.scale.com/leaderboard/mcp_atlas) Source scale Original model label claude-haiku-4-5 Observed 2026-10-22T03:30:07.875Z Evidence SHA-256 `7b0bc178e35a1688340e9efca5795a8f54d07e56a98f9cacd5e4471e0da266e9` Source locator `/rows/33` |  |  |  |  |  |  |
| [DesignArena Browser game generation](https://benchize.com/benchmarks/designarena%3Agame-dev.md) | 1,097elo | — | 85 / 117 | — | [designarena · public](https://www.designarena.ai/api/leaderboard#board=game-dev) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1097 \| elo \| game-dev \| Higher is better \| ### Configuration arenaType models category gamedev inputModality Not reported sourceModelVersion claude-haiku-4-5 variationName public ### Reported statistics averageGenerationTimeMs 47201 battles 4502 losses 2490 standardError Not reported winRate 44.7 wins 2012 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=game-dev) Source designarena Original model label Claude Haiku 4.5 Source version public Observed 2026-10-21T16:38:01.100Z Evidence SHA-256 `53db6a89112ce8cdf644568c666d4b8309aece764cfcedb291ffd52ae3934e30` Source locator `/data/126` |  |  |  |  |  |  |
| [DesignArena UI component generation](https://benchize.com/benchmarks/designarena%3Aui-components.md) | 1,100elo | — | 84 / 112 | — | [designarena · public](https://www.designarena.ai/api/leaderboard#board=ui-components) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1100 \| elo \| ui-components \| Higher is better \| ### Configuration arenaType models category uicomponent inputModality Not reported sourceModelVersion claude-haiku-4-5 variationName public ### Reported statistics averageGenerationTimeMs 37987 battles 1865 losses 1070 standardError Not reported winRate 42.6 wins 795 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=ui-components) Source designarena Original model label Claude Haiku 4.5 Source version public Observed 2026-10-21T16:37:57.993Z Evidence SHA-256 `46482ec1b4012638eace65493daea4750e2088098018fd9d5fc46126fc1f8557` Source locator `/data/125` |  |  |  |  |  |  |
| [DesignArena ASCII art generation](https://benchize.com/benchmarks/designarena%3Aascii.md) | 1,140elo | — | 43 / 65 | — | [designarena · public](https://www.designarena.ai/api/leaderboard#board=ascii) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1140 \| elo \| ascii \| Higher is better \| ### Configuration arenaType models category ascii inputModality Not reported sourceModelVersion claude-haiku-4-5 variationName public ### Reported statistics averageGenerationTimeMs 11177 battles 353 losses 179 standardError Not reported winRate 49.3 wins 174 ### Source [Original source](https://www.designarena.ai/api/leaderboard#board=ascii) Source designarena Original model label Claude Haiku 4.5 Source version public Observed 2026-10-21T16:38:03.513Z Evidence SHA-256 `0603b8c65b44ee887f5b566feb469465f59635137a6ac5c6d94c23bcf3776311` Source locator `/data/61` |  |  |  |  |  |  |

---
Cite as: Claude Haiku 4.5 — Benchize, https://benchize.com/models/claude-haiku-4-5 (retrieved 2026-10-22)