As of October 20, 2026, Anthropic Claude Haiku 4.5 ranks #112 of 213 models overall with 16.9 on the Artificial Analysis Intelligence Index. Its strongest category is coding, where it ranks #77 of 136. Anthropic Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens. It was released on October 15, 2025.
Its output price ranks 258 of 349 priced models, counting from the cheapest.
Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance...
anthropic/claude-haiku-4.5 The tick on a coding bar is the median of every model measured on it. Outside coding, the place is this model's rank on that board. A board that never ran the model says so.
Measured in 6 of 6 categories. A place counts only the models measured on that board.
The index averages only the coding benchmarks that have a result; the others are skipped, never filled in. How it is weighted
Not measured on SWE-bench Lite, SWE-bench (full test set), SWE-bench Multimodal, Terminal-Bench 4 (Vals AI), IOI (Vals AI), SWE-bench Multimodal (Vals AI), Vibe Code Bench 1-100 (Vals AI), SkillsBench (Vals AI), CyberBench (Vals AI), LiveCodeBench (Artificial Analysis run), SciCode, Terminal-Bench Hard (Artificial Analysis run), DeepSWE, FrontierCode, FrontierSWE, GSO-Bench, MirrorCode, Cybench, ALE-Bench, LMArena WebDev: from an image
Not measured on Design Arena: agents building web apps, Design Arena: agents building full-stack apps, Design Arena: agents building mobile apps, Design Arena: agents building Android apps, Design Arena: agents building Godot games, Design Arena: agents building games, Design Arena: HTML slides, Design Arena: PowerPoint slides, Design Arena: spreadsheets, SREBench (Vals AI), Harvey's Legal Agent Benchmark (Vals AI), BioMysteryBench (Vals AI), Terminal-Bench-Science (Vals AI), Vals Science Index, Time Horizon Index (Vals AI), CUA-bench (Vals AI), Robo-bench (Vals AI), τ²-Bench Telecom, APEX-Agents, METR time horizons, Remote Labor Index, GDPval, OSWorld 2.0, PostTrainBench, TheAgentCompany, OSWorld, SpireClimb, LMArena Search
Not measured on MysteryMechanism (Vals AI), GPQA Diamond (Artificial Analysis run), Humanity's Last Exam (Artificial Analysis run), Humanity's Last Exam, SimpleBench, Mystery game puzzles, EBR-bench, VPCT, BIG-Bench Hard, LMArena Vision: diagrams, LMArena Vision: homework
Not measured on MATH 500 (Vals AI), AA Math Index, AIME 2025 (Artificial Analysis run), FrontierMath, tiers 1 to 3, FrontierMath, tier 4, ProofBench, GSM8K
Not measured on MMLU Pro (Artificial Analysis run), MMLU, GeoBench, AI2 Reasoning Challenge
Not measured on Vals Index, IFBench, Long context reasoning (AA-LCR), Fiction.LiveBench, Creative writing (Lech Mazur), CL-bench, Surface Evolver Bench, Furniture assembly, HellaSwag, WinoGrande, PIQA, CL-bench Life, LMArena Vision, LMArena Vision: English, LMArena Vision: Chinese, LMArena Vision: reading text, LMArena Vision: humor, LMArena Vision: captioning, LMArena Vision: recognizing things, LMArena Vision: creative writing
Not measured on Design Arena: images, Design Arena: graphic design, Design Arena: logos, LMArena Text to Image, LMArena Image Edit
Public boards shown as published (Vals AI, Artificial Analysis, Design Arena) and kept out of the coding index. Each links to its source on the benchmark page.
| Benchmark | Result | Index weight |
|---|---|---|
| Aider polyglotCoding | Not measured | 35% |
| SWE-bench VerifiedCoding | 66.6% | 30% |
| SWE-bench LiteCoding | Not measured | Not in index |
| SWE-bench (full test set)Coding | Not measured | Not in index |
| SWE-bench MultimodalCoding | Not measured | Not in index |
| SWE-bench MultilingualCoding | 64.7% | Not in index |
| LiveBench codingCoding | 48.1% | 20% |
| LMArena WebDevCoding | 1330 | 15% |
| Vals IndexGeneral | Not measured | Not in index |
| GPQA DiamondReasoning | 72.2% | Not in index |
| AIMEMath | 82.7% | Not in index |
| MMLU ProKnowledge | 78.7% | Not in index |
| AA Intelligence IndexIntelligence | 16.9 | Not in index |
| AA Coding IndexCoding | 43.9 | Not in index |
| AA Agentic IndexAgentic work | 8.0 | Not in index |
| Design Arena: websitesWeb design | 1122 | Not in index |
| Design Arena: UI componentsWeb design | 1100 | Not in index |
| Design Arena: game devWeb design | 1097 | Not in index |
| Design Arena: data vizWeb design | 1124 | Not in index |
| Design Arena: 3DWeb design | 1072 | Not in index |
| Design Arena: all code categoriesWeb design | 1115 | Not in index |
| Design Arena: SVGWeb design | 1025 | Not in index |
| Design Arena: ASCII artWeb design | 1140 | Not in index |
| Design Arena: agents building web appsAgentic work | Not measured | Not in index |
| Design Arena: agents building full-stack appsAgentic work | Not measured | Not in index |
| Design Arena: agents building mobile appsAgentic work | Not measured | Not in index |
| Design Arena: agents building Android appsAgentic work | Not measured | Not in index |
| Design Arena: agents building Godot gamesAgentic work | Not measured | Not in index |
| Design Arena: agents building gamesAgentic work | Not measured | Not in index |
| Design Arena: HTML slidesAgentic work | Not measured | Not in index |
| Design Arena: PowerPoint slidesAgentic work | Not measured | Not in index |
| Design Arena: spreadsheetsAgentic work | Not measured | Not in index |
| Design Arena: imagesImage generation | Not measured | Not in index |
| Design Arena: graphic designImage generation | Not measured | Not in index |
| Design Arena: logosImage generation | Not measured | Not in index |
| LiveBench reasoningReasoning | 61.7% | Not in index |
| LiveBench data analysisGeneral | 59.3% | Not in index |
| LiveBench researchAgentic work | 42.2% | Not in index |
| LiveBench browserAgentic work | 40.0% | Not in index |
| LiveBench long agenticAgentic work | 11.1% | Not in index |
| LiveBench designWeb design | 59.9% | Not in index |
| LiveCodeBench (Vals AI)Coding | 41.2% | Not in index |
| SWE-bench Verified (Vals AI)Coding | 66.6% | Not in index |
| Terminal-Bench 4 (Vals AI)Coding | Not measured | Not in index |
| Vibe Code Bench (Vals AI)Coding | 11.4% | Not in index |
| Code migration (Vals AI)Coding | 7.5% | Not in index |
| ProgramBench (Vals AI)Coding | 0.0% | Not in index |
| SREBench (Vals AI)Agentic work | Not measured | Not in index |
| IOI (Vals AI)Coding | Not measured | Not in index |
| Terminal-Bench 2.1 (Vals AI)Coding | 43.8% | Not in index |
| SWE-bench Multimodal (Vals AI)Coding | Not measured | Not in index |
| Vibe Code Bench 1-100 (Vals AI)Coding | Not measured | Not in index |
| SkillsBench (Vals AI)Coding | Not measured | Not in index |
| CyberBench (Vals AI)Coding | Not measured | Not in index |
| Finance Agent (Vals AI)Agentic work | 31.0% | Not in index |
| Tax Agent Bench (Vals AI)Agentic work | 5.9% | Not in index |
| Legal Research Bench (Vals AI)Agentic work | 10.6% | Not in index |
| Harvey's Legal Agent Benchmark (Vals AI)Agentic work | Not measured | Not in index |
| Excel Modeling Benchmark (Vals AI)Agentic work | 22.7% | Not in index |
| BioMysteryBench (Vals AI)Agentic work | Not measured | Not in index |
| Terminal-Bench-Science (Vals AI)Agentic work | Not measured | Not in index |
| Vals Science IndexAgentic work | Not measured | Not in index |
| Time Horizon Index (Vals AI)Agentic work | Not measured | Not in index |
| CUA-bench (Vals AI)Agentic work | Not measured | Not in index |
| Robo-bench (Vals AI)Agentic work | Not measured | Not in index |
| Public Benefits Bench (Vals AI)General | 54.3% | Not in index |
| MysteryMechanism (Vals AI)Reasoning | Not measured | Not in index |
| MMMU Pro (Vals AI)Reasoning | 46.1% | Not in index |
| MATH 500 (Vals AI)Math | Not measured | Not in index |
| MGSM (Vals AI)Math | 92.2% | Not in index |
| SAGE (Vals AI)General | 31.8% | Not in index |
| LegalBench (Vals AI)Knowledge | 81.2% | Not in index |
| CaseLaw (Vals AI)Knowledge | 56.5% | Not in index |
| CorpFin (Vals AI)Knowledge | 60.6% | Not in index |
| TaxEval (Vals AI)Knowledge | 67.5% | Not in index |
| MortgageTax (Vals AI)Knowledge | 62.2% | Not in index |
| MedQA (Vals AI)Knowledge | 79.6% | Not in index |
| MedCode (Vals AI)Knowledge | 32.7% | Not in index |
| MedScribe (Vals AI)General | 85.2% | Not in index |
| Vals Multimodal IndexGeneral | 42.9% | Not in index |
| AA Math IndexMath | Not measured | Not in index |
| GPQA Diamond (Artificial Analysis run)Reasoning | Not measured | Not in index |
| Humanity's Last Exam (Artificial Analysis run)Reasoning | Not measured | Not in index |
| MMLU Pro (Artificial Analysis run)Knowledge | Not measured | Not in index |
| LiveCodeBench (Artificial Analysis run)Coding | Not measured | Not in index |
| SciCodeCoding | Not measured | Not in index |
| Terminal-Bench Hard (Artificial Analysis run)Coding | Not measured | Not in index |
| IFBenchGeneral | Not measured | Not in index |
| Long context reasoning (AA-LCR)General | Not measured | Not in index |
| τ²-Bench TelecomAgentic work | Not measured | Not in index |
| AIME 2025 (Artificial Analysis run)Math | Not measured | Not in index |
| Terminal-BenchCoding | 35.5% | Not in index |
| DeepSWECoding | Not measured | Not in index |
| FrontierCodeCoding | Not measured | Not in index |
| FrontierSWECoding | Not measured | Not in index |
| GSO-BenchCoding | Not measured | Not in index |
| WeirdMLCoding | 45.4% | Not in index |
| MirrorCodeCoding | Not measured | Not in index |
| APEX-AgentsAgentic work | Not measured | Not in index |
| METR time horizonsAgentic work | Not measured | Not in index |
| Remote Labor IndexAgentic work | Not measured | Not in index |
| GDPvalAgentic work | Not measured | Not in index |
| OSWorld 2.0Agentic work | Not measured | Not in index |
| BALROGAgentic work | 31.2% | Not in index |
| DeepResearch BenchAgentic work | 45.5% | Not in index |
| PostTrainBenchAgentic work | Not measured | Not in index |
| GPQA Diamond (Epoch AI run)Reasoning | 71.2% | Not in index |
| ARC-AGIReasoning | 47.7% | Not in index |
| ARC-AGI-2Reasoning | 4.0% | Not in index |
| Humanity's Last ExamReasoning | Not measured | Not in index |
| SimpleBenchReasoning | Not measured | Not in index |
| Chess puzzlesReasoning | 8.0% | Not in index |
| Mystery game puzzlesReasoning | Not measured | Not in index |
| EBR-benchReasoning | Not measured | Not in index |
| OTIS Mock AIMEMath | 66.7% | Not in index |
| FrontierMath, tiers 1 to 3Math | Not measured | Not in index |
| FrontierMath, tier 4Math | Not measured | Not in index |
| ProofBenchMath | Not measured | Not in index |
| MATH level 5Math | 96.4% | Not in index |
| GSM8KMath | Not measured | Not in index |
| SimpleQA VerifiedKnowledge | 13.2% | Not in index |
| MMLUKnowledge | Not measured | Not in index |
| GeoBenchKnowledge | Not measured | Not in index |
| DTBenchGeneral | 73.6% | Not in index |
| LMCAGeneral | 30.9% | Not in index |
| Fiction.LiveBenchGeneral | Not measured | Not in index |
| Creative writing (Lech Mazur)General | Not measured | Not in index |
| CL-benchGeneral | Not measured | Not in index |
| Surface Evolver BenchGeneral | Not measured | Not in index |
| Furniture assemblyGeneral | Not measured | Not in index |
| ExploitBenchGeneral | 13.7% | Not in index |
| CybenchCoding | Not measured | Not in index |
| TheAgentCompanyAgentic work | Not measured | Not in index |
| OSWorldAgentic work | Not measured | Not in index |
| SpireClimbAgentic work | Not measured | Not in index |
| VPCTReasoning | Not measured | Not in index |
| BIG-Bench HardReasoning | Not measured | Not in index |
| AI2 Reasoning ChallengeKnowledge | Not measured | Not in index |
| HellaSwagGeneral | Not measured | Not in index |
| WinoGrandeGeneral | Not measured | Not in index |
| PIQAGeneral | Not measured | Not in index |
| CL-bench LifeGeneral | Not measured | Not in index |
| ALE-BenchCoding | Not measured | Not in index |
| Vending-Bench 2Agentic work | $459 | Not in index |
| Epoch Capabilities IndexIntelligence | 142.4 | Not in index |
| LMArena TextGeneral | 1397 | Not in index |
| LMArena Text: codingCoding | 1454 | Not in index |
| LMArena Text: hard promptsReasoning | 1420 | Not in index |
| LMArena Text: mathMath | 1397 | Not in index |
| LMArena Text: instruction followingGeneral | 1414 | Not in index |
| LMArena Text: creative writingGeneral | 1371 | Not in index |
| LMArena VisionGeneral | Not measured | Not in index |
| LMArena Text: EnglishGeneral | 1414 | Not in index |
| LMArena Text: non-EnglishGeneral | 1377 | Not in index |
| LMArena Text: ChineseGeneral | 1418 | Not in index |
| LMArena Text: GermanGeneral | 1374 | Not in index |
| LMArena Text: JapaneseGeneral | 1338 | Not in index |
| LMArena Text: KoreanGeneral | 1346 | Not in index |
| LMArena Text: FrenchGeneral | 1409 | Not in index |
| LMArena Text: SpanishGeneral | 1421 | Not in index |
| LMArena Text: RussianGeneral | 1381 | Not in index |
| LMArena Text: PolishGeneral | 1342 | Not in index |
| LMArena Text: multi-turnGeneral | 1409 | Not in index |
| LMArena Text: longer queriesGeneral | 1427 | Not in index |
| LMArena Text: expertReasoning | 1443 | Not in index |
| LMArena Text: hard prompts in EnglishReasoning | 1434 | Not in index |
| LMArena Text: software and ITCoding | 1439 | Not in index |
| LMArena Text: mathematical workMath | 1430 | Not in index |
| LMArena Text: scienceKnowledge | 1406 | Not in index |
| LMArena Text: medicine and healthcareKnowledge | 1394 | Not in index |
| LMArena Text: legal and governmentKnowledge | 1400 | Not in index |
| LMArena Text: business and financeKnowledge | 1400 | Not in index |
| LMArena Text: writing and languageGeneral | 1387 | Not in index |
| LMArena Text: entertainment and mediaGeneral | 1369 | Not in index |
| LMArena WebDev: ReactCoding | 1326 | Not in index |
| LMArena WebDev: HTMLCoding | 1318 | Not in index |
| LMArena WebDev: from an imageCoding | Not measured | Not in index |
| LMArena Vision: EnglishGeneral | Not measured | Not in index |
| LMArena Vision: ChineseGeneral | Not measured | Not in index |
| LMArena Vision: reading textGeneral | Not measured | Not in index |
| LMArena Vision: diagramsReasoning | Not measured | Not in index |
| LMArena Vision: homeworkReasoning | Not measured | Not in index |
| LMArena Vision: humorGeneral | Not measured | Not in index |
| LMArena Vision: captioningGeneral | Not measured | Not in index |
| LMArena Vision: recognizing thingsGeneral | Not measured | Not in index |
| LMArena Vision: creative writingGeneral | Not measured | Not in index |
| LMArena SearchAgentic work | Not measured | Not in index |
| LMArena DocumentGeneral | 1420 | Not in index |
| LMArena Text to ImageImage generation | Not measured | Not in index |
| LMArena Image EditImage generation | Not measured | Not in index |
A list price says what a token costs. The cost per solved task says what a working answer costs, which is the number that decides.
Beside a result, many boards publish what the run cost, how long it took, the tokens it spent or the votes behind a rating. Each figure as its board printed it, one board per row.
| Board | Result | What the run cost | Also published |
|---|---|---|---|
| SWE-bench Verified | 66.6% | $0.3309 per attempt | Nothing else |
| SWE-bench Multilingual | 64.7% | $0.3795 per attempt | Nothing else |
| LiveBench coding | 48.1% | $0.494 per task | Output tokens per task 19,224 |
| LiveBench research | 42.2% | $0.4322 per task | Output tokens per task 14,045 |
| LiveBench browser | 40.0% | $1.5018 per task | Output tokens per task 24,460 |
| LiveBench long agentic | 11.1% | $2.5485 per task | Output tokens per task 77,255 |
| LiveBench design | 59.9% | $0.1558 per task | Output tokens per task 11,478 |
| SWE-bench Verified (Vals AI) | 66.6% | $0.37 per test | Duration, as printed 4m 02s |
| Vibe Code Bench (Vals AI) | 11.4% | $1.31 per test | Duration, as printed 12m 56s |
| Code migration (Vals AI) | 7.5% | $0.98 per test | Duration, as printed 9m 54s |
| Terminal-Bench 2.1 (Vals AI) | 43.8% | $0.36 per test | Duration, as printed 10m 49s |
| Finance Agent (Vals AI) | 31.0% | $0.60 per test | Duration, as printed 2m 57s |
| Tax Agent Bench (Vals AI) | 5.9% | $0.23 per test | Duration, as printed 2m 35s |
|---|---|---|---|
| Legal Research Bench (Vals AI) | 10.6% | $0.27 per test | Duration, as printed 4m 03s |
| Excel Modeling Benchmark (Vals AI) | 22.7% | $1.03 per test | Duration, as printed 11m 59s |
| Public Benefits Bench (Vals AI) | 54.3% | $0.22 per test | Nothing else |
| Vals Multimodal Index | 42.9% | $0.41 per test | Duration, as printed 4m 43s |
| WeirdML | 45.4% | $0.0894 per run | Median lines of code 174 |
| ARC-AGI | 47.7% | $0.2558 per task | Nothing else |
| ARC-AGI-2 | 4.0% | $0.3766 per task | Nothing else |
| LMArena WebDev | 1330 Elo | Not published | Votes 29,061 · Interval (±) ±4.9 |
| GPQA Diamond | 72.2% | Not published | Latency, as printed 1m 23s |
| AIME | 82.7% | Not published | Latency, as printed 2m 38s |
| MMLU Pro | 78.7% | Not published | Latency, as printed 26 s |
| Design Arena: websites | 1122 Elo | Not published | Battles 23,935 |
| Design Arena: UI components | 1100 Elo | Not published | Battles 1,865 |
| Design Arena: game dev | 1097 Elo | Not published | Battles 4,502 |
| Design Arena: data viz | 1124 Elo | Not published | Battles 1,131 |
| Design Arena: 3D | 1072 Elo | Not published | Battles 6,343 |
| Design Arena: all code categories | 1115 Elo | Not published | Battles 35,541 |
| Design Arena: SVG | 1025 Elo | Not published | Battles 4,714 |
| Design Arena: ASCII art | 1140 Elo | Not published | Battles 353 |
| LiveCodeBench (Vals AI) | 41.2% | Not published | Latency, as printed 1m 22s |
| ProgramBench (Vals AI) | 0.0% | Not published | Duration, as printed 7m 37s |
| MMMU Pro (Vals AI) | 46.1% | Not published | Latency, as printed 40 s |
| MGSM (Vals AI) | 92.2% | Not published | Latency, as printed 9.6 s |
| SAGE (Vals AI) | 31.8% | Not published | Latency, as printed 1m 06s |
| LegalBench (Vals AI) | 81.2% | Not published | Latency, as printed 10 s |
| CaseLaw (Vals AI) | 56.5% | Not published | Latency, as printed 33 s |
| CorpFin (Vals AI) | 60.6% | Not published | Latency, as printed 14 s |
| TaxEval (Vals AI) | 67.5% | Not published | Latency, as printed 48 s |
| MortgageTax (Vals AI) | 62.2% | Not published | Latency, as printed 30 s |
| MedQA (Vals AI) | 79.6% | Not published | Latency, as printed 24 s |
| MedCode (Vals AI) | 32.7% | Not published | Latency, as printed 34 s |
| MedScribe (Vals AI) | 85.2% | Not published | Latency, as printed 1m 06s |
| LMArena Text | 1397 Elo | Not published | Votes 149,199 · Interval (±) ±2.4 |
| LMArena Text: coding | 1454 Elo | Not published | Votes 38,427 · Interval (±) ±4.0 |
| LMArena Text: hard prompts | 1420 Elo | Not published | Votes 92,034 · Interval (±) ±3.0 |
| LMArena Text: math | 1397 Elo | Not published | Votes 8,311 · Interval (±) ±7.0 |
| LMArena Text: instruction following | 1414 Elo | Not published | Votes 47,790 · Interval (±) ±3.7 |
| LMArena Text: creative writing | 1371 Elo | Not published | Votes 25,394 · Interval (±) ±4.7 |
| LMArena Text: English | 1414 Elo | Not published | Votes 64,734 · Interval (±) ±3.3 |
| LMArena Text: non-English | 1377 Elo | Not published | Votes 84,413 · Interval (±) ±3.0 |
| LMArena Text: Chinese | 1418 Elo | Not published | Votes 10,532 · Interval (±) ±6.4 |
| LMArena Text: German | 1374 Elo | Not published | Votes 2,468 · Interval (±) ±13.2 |
| LMArena Text: Japanese | 1338 Elo | Not published | Votes 1,712 · Interval (±) ±16.4 |
| LMArena Text: Korean | 1346 Elo | Not published | Votes 2,378 · Interval (±) ±13.6 |
| LMArena Text: French | 1409 Elo | Not published | Votes 4,213 · Interval (±) ±11.1 |
| LMArena Text: Spanish | 1421 Elo | Not published | Votes 4,363 · Interval (±) ±10.3 |
| LMArena Text: Russian | 1381 Elo | Not published | Votes 16,646 · Interval (±) ±5.2 |
| LMArena Text: Polish | 1342 Elo | Not published | Votes 3,483 · Interval (±) ±10.6 |
| LMArena Text: multi-turn | 1409 Elo | Not published | Votes 25,888 · Interval (±) ±4.5 |
| LMArena Text: longer queries | 1427 Elo | Not published | Votes 56,594 · Interval (±) ±3.7 |
| LMArena Text: expert | 1443 Elo | Not published | Votes 13,709 · Interval (±) ±5.9 |
| LMArena Text: hard prompts in English | 1434 Elo | Not published | Votes 40,648 · Interval (±) ±3.9 |
| LMArena Text: software and IT | 1439 Elo | Not published | Votes 56,191 · Interval (±) ±3.5 |
| LMArena Text: mathematical work | 1430 Elo | Not published | Votes 7,959 · Interval (±) ±7.3 |
| LMArena Text: science | 1406 Elo | Not published | Votes 24,337 · Interval (±) ±4.6 |
| LMArena Text: medicine and healthcare | 1394 Elo | Not published | Votes 10,077 · Interval (±) ±6.7 |
| LMArena Text: legal and government | 1400 Elo | Not published | Votes 11,207 · Interval (±) ±6.3 |
| LMArena Text: business and finance | 1400 Elo | Not published | Votes 28,782 · Interval (±) ±4.4 |
| LMArena Text: writing and language | 1387 Elo | Not published | Votes 36,448 · Interval (±) ±4.0 |
| LMArena Text: entertainment and media | 1369 Elo | Not published | Votes 32,219 · Interval (±) ±4.3 |
| LMArena WebDev: React | 1326 Elo | Not published | Votes 16,000 · Interval (±) ±6.0 |
| LMArena WebDev: HTML | 1318 Elo | Not published | Votes 10,648 · Interval (±) ±9.3 |
| LMArena Document | 1420 Elo | Not published | Votes 34,677 · Interval (±) ±5.9 |