As of October 20, 2026, Anthropic Claude Sonnet 4.6 ranks #50 of 213 models overall with 30.1 on the Artificial Analysis Intelligence Index. Its strongest category is knowledge, where it ranks #28 of 147. Anthropic Claude Sonnet 4.6 costs $3.00 per million input tokens and $15.00 per million output tokens. It was released on February 17, 2026.
Its output price ranks 315 of 349 priced models, counting from the cheapest.
Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. It excels at iterative development, complex codebase navigation, end-to-end project management with...
anthropic/claude-sonnet-4.6 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, SWE-bench Multilingual, Terminal-Bench 4 (Vals AI), IOI (Vals AI), SWE-bench Multimodal (Vals AI), Vibe Code Bench 1-100 (Vals AI), CyberBench (Vals AI), LiveCodeBench (Artificial Analysis run), FrontierSWE, GSO-Bench, MirrorCode, Cybench, ALE-Bench
Not measured on 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), METR time horizons, Remote Labor Index, GDPval, BALROG, PostTrainBench, TheAgentCompany, SpireClimb, LMArena Search
Not measured on MysteryMechanism (Vals AI), Humanity's Last Exam, SimpleBench, EBR-bench, VPCT, BIG-Bench Hard
Not measured on MATH 500 (Vals AI), MGSM (Vals AI), AA Math Index, AIME 2025 (Artificial Analysis run), FrontierMath, tiers 1 to 3, FrontierMath, tier 4, MATH level 5, GSM8K
Not measured on MedCode (Vals AI), MMLU Pro (Artificial Analysis run), MMLU, GeoBench, AI2 Reasoning Challenge
Not measured on Vals Index, MedScribe (Vals AI), Fiction.LiveBench, Creative writing (Lech Mazur), CL-bench, Surface Evolver Bench, Furniture assembly, HellaSwag, WinoGrande, PIQA, CL-bench Life, LMArena Vision: captioning
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 | Not measured | 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 | Not measured | Not in index |
| LiveBench codingCoding | 66.6% | 20% |
| LMArena WebDevCoding | 1523 | 15% |
| Vals IndexGeneral | Not measured | Not in index |
| GPQA DiamondReasoning | 85.6% | Not in index |
| AIMEMath | 92.3% | Not in index |
| MMLU ProKnowledge | 87.3% | Not in index |
| AA Intelligence IndexIntelligence | 30.1 | Not in index |
| AA Coding IndexCoding | 63.0 | Not in index |
| AA Agentic IndexAgentic work | 31.8 | Not in index |
| Design Arena: websitesWeb design | 1285 | Not in index |
| Design Arena: UI componentsWeb design | 1272 | Not in index |
| Design Arena: game devWeb design | 1259 | Not in index |
| Design Arena: data vizWeb design | 1278 | Not in index |
| Design Arena: 3DWeb design | 1234 | Not in index |
| Design Arena: all code categoriesWeb design | 1276 | Not in index |
| Design Arena: SVGWeb design | 1196 | Not in index |
| Design Arena: ASCII artWeb design | 1236 | Not in index |
| Design Arena: agents building web appsAgentic work | 1190 | Not in index |
| Design Arena: agents building full-stack appsAgentic work | 1213 | Not in index |
| Design Arena: agents building mobile appsAgentic work | 1211 | Not in index |
| Design Arena: agents building Android appsAgentic work | 1212 | Not in index |
| Design Arena: agents building Godot gamesAgentic work | 1226 | Not in index |
| Design Arena: agents building gamesAgentic work | 1177 | 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 | 86.4% | Not in index |
| LiveBench data analysisGeneral | 76.1% | Not in index |
| LiveBench researchAgentic work | 63.6% | Not in index |
| LiveBench browserAgentic work | 68.0% | Not in index |
| LiveBench long agenticAgentic work | 23.1% | Not in index |
| LiveBench designWeb design | 71.8% | Not in index |
| LiveCodeBench (Vals AI)Coding | 82.1% | Not in index |
| SWE-bench Verified (Vals AI)Coding | 77.4% | Not in index |
| Terminal-Bench 4 (Vals AI)Coding | Not measured | Not in index |
| Vibe Code Bench (Vals AI)Coding | 55.8% | Not in index |
| Code migration (Vals AI)Coding | 39.9% | Not in index |
| ProgramBench (Vals AI)Coding | 0.5% | 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 | 57.3% | 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 | 49.0% | Not in index |
| CyberBench (Vals AI)Coding | Not measured | Not in index |
| Finance Agent (Vals AI)Agentic work | 51.0% | Not in index |
| Tax Agent Bench (Vals AI)Agentic work | 25.8% | Not in index |
| Legal Research Bench (Vals AI)Agentic work | 38.5% | Not in index |
| Harvey's Legal Agent Benchmark (Vals AI)Agentic work | Not measured | Not in index |
| Excel Modeling Benchmark (Vals AI)Agentic work | 60.2% | 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 | 62.5% | Not in index |
| MysteryMechanism (Vals AI)Reasoning | Not measured | Not in index |
| MMMU Pro (Vals AI)Reasoning | 83.6% | Not in index |
| MATH 500 (Vals AI)Math | Not measured | Not in index |
| MGSM (Vals AI)Math | Not measured | Not in index |
| SAGE (Vals AI)General | 46.6% | Not in index |
| LegalBench (Vals AI)Knowledge | 82.1% | Not in index |
| CaseLaw (Vals AI)Knowledge | 64.0% | Not in index |
| CorpFin (Vals AI)Knowledge | 65.3% | Not in index |
| TaxEval (Vals AI)Knowledge | 77.1% | Not in index |
| MortgageTax (Vals AI)Knowledge | 67.7% | Not in index |
| MedQA (Vals AI)Knowledge | 92.1% | Not in index |
| MedCode (Vals AI)Knowledge | Not measured | Not in index |
| MedScribe (Vals AI)General | Not measured | Not in index |
| Vals Multimodal IndexGeneral | 60.6% | Not in index |
| AA Math IndexMath | Not measured | Not in index |
| GPQA Diamond (Artificial Analysis run)Reasoning | 87.5% | Not in index |
| Humanity's Last Exam (Artificial Analysis run)Reasoning | 33.6% | 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 | 50.1% | Not in index |
| Terminal-Bench Hard (Artificial Analysis run)Coding | 53.0% | Not in index |
| IFBenchGeneral | 56.6% | Not in index |
| Long context reasoning (AA-LCR)General | 80.0% | Not in index |
| τ²-Bench TelecomAgentic work | 79.5% | Not in index |
| AIME 2025 (Artificial Analysis run)Math | Not measured | Not in index |
| Terminal-BenchCoding | 53.4% | Not in index |
| DeepSWECoding | 29.9% | Not in index |
| FrontierCodeCoding | 24.3% | Not in index |
| FrontierSWECoding | Not measured | Not in index |
| GSO-BenchCoding | Not measured | Not in index |
| WeirdMLCoding | 66.1% | Not in index |
| MirrorCodeCoding | Not measured | Not in index |
| APEX-AgentsAgentic work | 43.0% | 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 | 9.3% | Not in index |
| BALROGAgentic work | Not measured | Not in index |
| DeepResearch BenchAgentic work | 54.9% | Not in index |
| PostTrainBenchAgentic work | Not measured | Not in index |
| GPQA Diamond (Epoch AI run)Reasoning | 87.4% | Not in index |
| ARC-AGIReasoning | 86.5% | Not in index |
| ARC-AGI-2Reasoning | 60.4% | Not in index |
| Humanity's Last ExamReasoning | Not measured | Not in index |
| SimpleBenchReasoning | Not measured | Not in index |
| Chess puzzlesReasoning | 23.0% | Not in index |
| Mystery game puzzlesReasoning | 16.0% | Not in index |
| EBR-benchReasoning | Not measured | Not in index |
| OTIS Mock AIMEMath | 85.8% | Not in index |
| FrontierMath, tiers 1 to 3Math | Not measured | Not in index |
| FrontierMath, tier 4Math | Not measured | Not in index |
| ProofBenchMath | 45.0% | Not in index |
| MATH level 5Math | Not measured | Not in index |
| GSM8KMath | Not measured | Not in index |
| SimpleQA VerifiedKnowledge | 35.5% | Not in index |
| MMLUKnowledge | Not measured | Not in index |
| GeoBenchKnowledge | Not measured | Not in index |
| DTBenchGeneral | 89.9% | Not in index |
| LMCAGeneral | 46.5% | 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 | 23.6% | Not in index |
| CybenchCoding | Not measured | Not in index |
| TheAgentCompanyAgentic work | Not measured | Not in index |
| OSWorldAgentic work | 72.1% | 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 | $7,204 | Not in index |
| Epoch Capabilities IndexIntelligence | 152.4 | Not in index |
| LMArena TextGeneral | 1458 | Not in index |
| LMArena Text: codingCoding | 1504 | Not in index |
| LMArena Text: hard promptsReasoning | 1485 | Not in index |
| LMArena Text: mathMath | 1463 | Not in index |
| LMArena Text: instruction followingGeneral | 1476 | Not in index |
| LMArena Text: creative writingGeneral | 1435 | Not in index |
| LMArena VisionGeneral | 1283 | Not in index |
| LMArena Text: EnglishGeneral | 1472 | Not in index |
| LMArena Text: non-EnglishGeneral | 1440 | Not in index |
| LMArena Text: ChineseGeneral | 1492 | Not in index |
| LMArena Text: GermanGeneral | 1428 | Not in index |
| LMArena Text: JapaneseGeneral | 1420 | Not in index |
| LMArena Text: KoreanGeneral | 1411 | Not in index |
| LMArena Text: FrenchGeneral | 1465 | Not in index |
| LMArena Text: SpanishGeneral | 1464 | Not in index |
| LMArena Text: RussianGeneral | 1439 | Not in index |
| LMArena Text: PolishGeneral | 1442 | Not in index |
| LMArena Text: multi-turnGeneral | 1464 | Not in index |
| LMArena Text: longer queriesGeneral | 1479 | Not in index |
| LMArena Text: expertReasoning | 1500 | Not in index |
| LMArena Text: hard prompts in EnglishReasoning | 1494 | Not in index |
| LMArena Text: software and ITCoding | 1491 | Not in index |
| LMArena Text: mathematical workMath | 1483 | Not in index |
| LMArena Text: scienceKnowledge | 1475 | Not in index |
| LMArena Text: medicine and healthcareKnowledge | 1451 | Not in index |
| LMArena Text: legal and governmentKnowledge | 1470 | Not in index |
| LMArena Text: business and financeKnowledge | 1464 | Not in index |
| LMArena Text: writing and languageGeneral | 1444 | Not in index |
| LMArena Text: entertainment and mediaGeneral | 1431 | Not in index |
| LMArena WebDev: ReactCoding | 1512 | Not in index |
| LMArena WebDev: HTMLCoding | 1527 | Not in index |
| LMArena WebDev: from an imageCoding | 1529 | Not in index |
| LMArena Vision: EnglishGeneral | 1298 | Not in index |
| LMArena Vision: ChineseGeneral | 1307 | Not in index |
| LMArena Vision: reading textGeneral | 1296 | Not in index |
| LMArena Vision: diagramsReasoning | 1306 | Not in index |
| LMArena Vision: homeworkReasoning | 1311 | Not in index |
| LMArena Vision: humorGeneral | 1248 | Not in index |
| LMArena Vision: captioningGeneral | Not measured | Not in index |
| LMArena Vision: recognizing thingsGeneral | 1195 | Not in index |
| LMArena Vision: creative writingGeneral | 1272 | Not in index |
| LMArena SearchAgentic work | Not measured | Not in index |
| LMArena DocumentGeneral | 1482 | 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 |
|---|---|---|---|
| LiveBench coding | 66.6% | $2.6287 per task | Output tokens per task 37,618 |
| LiveBench reasoning | 86.4% | $0.313 per task | Output tokens per task 20,620 |
| LiveBench data analysis | 76.1% | $0.4169 per task | Output tokens per task 27,034 |
| LiveBench research | 63.6% | $3.9018 per task | Output tokens per task 35,125 |
| LiveBench browser | 68.0% | $3.3753 per task | Output tokens per task 22,502 |
| LiveBench long agentic | 23.1% | $20.4196 per task | Output tokens per task 310,212 |
| LiveBench design | 71.8% | $2.1386 per task | Output tokens per task 52,865 |
| SWE-bench Verified (Vals AI) | 77.4% | $1.30 per test | Duration, as printed 8m 32s |
| Vibe Code Bench (Vals AI) | 55.8% | $7.35 per test | Duration, as printed 49m 42s |
| Code migration (Vals AI) | 39.9% | $38.51 per test | Duration, as printed 1h 40m |
| Terminal-Bench 2.1 (Vals AI) | 57.3% | $0.57 per test | Duration, as printed 11m 44s |
| SkillsBench (Vals AI) | 49.0% | $1.38 per test | Duration, as printed 14m 20s |
| Finance Agent (Vals AI) | 51.0% | $2.41 per test | Duration, as printed 11m 36s |
|---|---|---|---|
| Tax Agent Bench (Vals AI) | 25.8% | $1.19 per test | Duration, as printed 11m 47s |
| Legal Research Bench (Vals AI) | 38.5% | $2.30 per test | Duration, as printed 16m 30s |
| Excel Modeling Benchmark (Vals AI) | 60.2% | $7.23 per test | Duration, as printed 54m 16s |
| Public Benefits Bench (Vals AI) | 62.5% | $1.11 per test | Nothing else |
| Vals Multimodal Index | 60.6% | $1.58 per test | Duration, as printed 8m 56s |
| DeepSWE | 29.9% | $5.5224 mean cost per task | Solved in 4 tries 56.6% · Output tokens per task 76,160 · Agent steps per task 134 |
| WeirdML | 66.1% | $2.8594 per run | Median lines of code 178 |
| OSWorld 2.0 | 9.3% | $1550.00 estimated cost | With partial credit 33.9% |
| ARC-AGI | 86.5% | $1.1908 per task | Nothing else |
| ARC-AGI-2 | 60.4% | $2.6966 per task | Nothing else |
| ProofBench | 45.0% | $2.1621 per test | Time per proof 1h 31m |
| LMArena WebDev | 1523 Elo | Not published | Votes 22,822 · Interval (±) ±5.2 |
| GPQA Diamond | 85.6% | Not published | Latency, as printed 6m 18s |
| AIME | 92.3% | Not published | Latency, as printed 5m 28s |
| MMLU Pro | 87.3% | Not published | Latency, as printed 55 s |
| Design Arena: websites | 1285 Elo | Not published | Battles 13,777 |
| Design Arena: UI components | 1272 Elo | Not published | Battles 2,167 |
| Design Arena: game dev | 1259 Elo | Not published | Battles 4,485 |
| Design Arena: data viz | 1278 Elo | Not published | Battles 1,902 |
| Design Arena: 3D | 1234 Elo | Not published | Battles 5,715 |
| Design Arena: all code categories | 1276 Elo | Not published | Battles 26,317 |
| Design Arena: SVG | 1196 Elo | Not published | Battles 4,025 |
| Design Arena: ASCII art | 1236 Elo | Not published | Battles 411 |
| Design Arena: agents building web apps | 1190 Elo | Not published | Battles 2,825 |
| Design Arena: agents building full-stack apps | 1213 Elo | Not published | Battles 2,030 |
| Design Arena: agents building mobile apps | 1211 Elo | Not published | Battles 1,558 |
| Design Arena: agents building Android apps | 1212 Elo | Not published | Battles 1,101 |
| Design Arena: agents building games | 1177 Elo | Not published | Battles 245 |
| LiveCodeBench (Vals AI) | 82.1% | Not published | Latency, as printed 4m 55s |
| ProgramBench (Vals AI) | 0.5% | Not published | Duration, as printed 1h 12m |
| MMMU Pro (Vals AI) | 83.6% | Not published | Latency, as printed 2m 12s |
| SAGE (Vals AI) | 46.6% | Not published | Latency, as printed 2m 43s |
| LegalBench (Vals AI) | 82.1% | Not published | Latency, as printed 22 s |
| CaseLaw (Vals AI) | 64.0% | Not published | Latency, as printed 36 s |
| CorpFin (Vals AI) | 65.3% | Not published | Latency, as printed 18 s |
| TaxEval (Vals AI) | 77.1% | Not published | Latency, as printed 2m 08s |
| MortgageTax (Vals AI) | 67.7% | Not published | Latency, as printed 30 s |
| MedQA (Vals AI) | 92.1% | Not published | Latency, as printed 35 s |
| APEX-Agents | 43.0% | Not published | Mean score 57.6% |
| LMArena Text | 1458 Elo | Not published | Votes 70,931 · Interval (±) ±3.5 |
| LMArena Text: coding | 1504 Elo | Not published | Votes 19,724 · Interval (±) ±5.5 |
| LMArena Text: hard prompts | 1485 Elo | Not published | Votes 46,229 · Interval (±) ±4.2 |
| LMArena Text: math | 1463 Elo | Not published | Votes 3,792 · Interval (±) ±10.0 |
| LMArena Text: instruction following | 1476 Elo | Not published | Votes 23,711 · Interval (±) ±5.2 |
| LMArena Text: creative writing | 1435 Elo | Not published | Votes 12,367 · Interval (±) ±6.5 |
| LMArena Vision | 1283 Elo | Not published | Votes 27,408 · Interval (±) ±6.1 |
| LMArena Text: English | 1472 Elo | Not published | Votes 31,965 · Interval (±) ±4.7 |
| LMArena Text: non-English | 1440 Elo | Not published | Votes 38,962 · Interval (±) ±4.3 |
| LMArena Text: Chinese | 1492 Elo | Not published | Votes 4,520 · Interval (±) ±9.4 |
| LMArena Text: German | 1428 Elo | Not published | Votes 1,221 · Interval (±) ±17.8 |
| LMArena Text: Japanese | 1420 Elo | Not published | Votes 801 · Interval (±) ±22.4 |
| LMArena Text: Korean | 1411 Elo | Not published | Votes 1,153 · Interval (±) ±18.5 |
| LMArena Text: French | 1465 Elo | Not published | Votes 2,284 · Interval (±) ±14.0 |
| LMArena Text: Spanish | 1464 Elo | Not published | Votes 2,168 · Interval (±) ±13.8 |
| LMArena Text: Russian | 1439 Elo | Not published | Votes 7,983 · Interval (±) ±7.3 |
| LMArena Text: Polish | 1442 Elo | Not published | Votes 1,412 · Interval (±) ±15.8 |
| LMArena Text: multi-turn | 1464 Elo | Not published | Votes 12,693 · Interval (±) ±6.4 |
| LMArena Text: longer queries | 1479 Elo | Not published | Votes 31,070 · Interval (±) ±4.9 |
| LMArena Text: expert | 1500 Elo | Not published | Votes 7,384 · Interval (±) ±7.8 |
| LMArena Text: hard prompts in English | 1494 Elo | Not published | Votes 21,356 · Interval (±) ±5.4 |
| LMArena Text: software and IT | 1491 Elo | Not published | Votes 27,745 · Interval (±) ±4.9 |
| LMArena Text: mathematical work | 1483 Elo | Not published | Votes 3,982 · Interval (±) ±10.1 |
| LMArena Text: science | 1475 Elo | Not published | Votes 11,651 · Interval (±) ±6.4 |
| LMArena Text: medicine and healthcare | 1451 Elo | Not published | Votes 5,437 · Interval (±) ±8.8 |
| LMArena Text: legal and government | 1470 Elo | Not published | Votes 5,707 · Interval (±) ±8.6 |
| LMArena Text: business and finance | 1464 Elo | Not published | Votes 14,063 · Interval (±) ±6.1 |
| LMArena Text: writing and language | 1444 Elo | Not published | Votes 17,602 · Interval (±) ±5.7 |
| LMArena Text: entertainment and media | 1431 Elo | Not published | Votes 15,933 · Interval (±) ±6.0 |
| LMArena WebDev: React | 1512 Elo | Not published | Votes 18,106 · Interval (±) ±5.7 |
| LMArena WebDev: HTML | 1527 Elo | Not published | Votes 2,897 · Interval (±) ±11.9 |
| LMArena WebDev: from an image | 1529 Elo | Not published | Votes 8,312 · Interval (±) ±8.8 |
| LMArena Vision: English | 1298 Elo | Not published | Votes 11,555 · Interval (±) ±8.3 |
| LMArena Vision: Chinese | 1307 Elo | Not published | Votes 1,736 · Interval (±) ±16.8 |
| LMArena Vision: reading text | 1296 Elo | Not published | Votes 19,839 · Interval (±) ±6.1 |
| LMArena Vision: diagrams | 1306 Elo | Not published | Votes 7,630 · Interval (±) ±8.4 |
| LMArena Vision: homework | 1311 Elo | Not published | Votes 3,700 · Interval (±) ±10.7 |
| LMArena Vision: humor | 1248 Elo | Not published | Votes 1,060 · Interval (±) ±19.2 |
| LMArena Vision: recognizing things | 1195 Elo | Not published | Votes 162 · Interval (±) ±45.6 |
| LMArena Vision: creative writing | 1272 Elo | Not published | Votes 1,696 · Interval (±) ±15.2 |
| LMArena Document | 1482 Elo | Not published | Votes 57,813 · Interval (±) ±5.9 |