As of October 20, 2026, MiniMax M3 ranks #53 of 213 models overall with 29.2 on the Artificial Analysis Intelligence Index. Its strongest category is reasoning, where it ranks #23 of 203. MiniMax M3 costs $0.30 per million input tokens and $1.20 per million output tokens. It was released on May 31, 2026.
Its output price ranks 147 of 349 priced models, counting from the cheapest.
MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...
minimax/minimax-m3 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, ProgramBench (Vals AI), IOI (Vals AI), LiveCodeBench (Artificial Analysis run), Terminal-Bench, DeepSWE, FrontierSWE, GSO-Bench, WeirdML, MirrorCode, Cybench
Not measured on Design Arena: agents building Godot games, 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, DeepResearch Bench, PostTrainBench, TheAgentCompany, OSWorld, SpireClimb, LMArena Search
Not measured on MysteryMechanism (Vals AI), ARC-AGI, ARC-AGI-2, Humanity's Last Exam, EBR-bench, VPCT, BIG-Bench Hard
Not measured on AIME, 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 CaseLaw (Vals AI), MedQA (Vals AI), MMLU Pro (Artificial Analysis run), SimpleQA Verified, MMLU, GeoBench, AI2 Reasoning Challenge
Not measured on Vals Index, Fiction.LiveBench, Creative writing (Lech Mazur), CL-bench, Furniture assembly, ExploitBench, 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 | 70.9% | 20% |
| LMArena WebDevCoding | 1482 | 15% |
| Vals IndexGeneral | Not measured | Not in index |
| GPQA DiamondReasoning | 92.7% | Not in index |
| AIMEMath | Not measured | Not in index |
| MMLU ProKnowledge | 84.2% | Not in index |
| AA Intelligence IndexIntelligence | 29.2 | Not in index |
| AA Coding IndexCoding | 58.6 | Not in index |
| AA Agentic IndexAgentic work | 29.5 | Not in index |
| Design Arena: websitesWeb design | 1258 | Not in index |
| Design Arena: UI componentsWeb design | 1247 | Not in index |
| Design Arena: game devWeb design | 1210 | Not in index |
| Design Arena: data vizWeb design | 1235 | Not in index |
| Design Arena: 3DWeb design | 1209 | Not in index |
| Design Arena: all code categoriesWeb design | 1247 | Not in index |
| Design Arena: SVGWeb design | 1161 | Not in index |
| Design Arena: ASCII artWeb design | 1161 | Not in index |
| Design Arena: agents building web appsAgentic work | 1198 | Not in index |
| Design Arena: agents building full-stack appsAgentic work | 1197 | Not in index |
| Design Arena: agents building mobile appsAgentic work | 1168 | Not in index |
| Design Arena: agents building Android appsAgentic work | 1173 | Not in index |
| Design Arena: agents building Godot gamesAgentic work | Not measured | Not in index |
| Design Arena: agents building gamesAgentic work | 1156 | Not in index |
| Design Arena: HTML slidesAgentic work | 1180 | Not in index |
| Design Arena: PowerPoint slidesAgentic work | 1205 | Not in index |
| Design Arena: spreadsheetsAgentic work | 1250 | 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 | 74.5% | Not in index |
| LiveBench data analysisGeneral | 76.2% | Not in index |
| LiveBench researchAgentic work | 68.0% | Not in index |
| LiveBench browserAgentic work | 8.0% | Not in index |
| LiveBench long agenticAgentic work | 12.4% | Not in index |
| LiveBench designWeb design | 66.2% | Not in index |
| LiveCodeBench (Vals AI)Coding | 82.2% | Not in index |
| SWE-bench Verified (Vals AI)Coding | 75.0% | Not in index |
| Terminal-Bench 4 (Vals AI)Coding | 1.0% | Not in index |
| Vibe Code Bench (Vals AI)Coding | 47.6% | Not in index |
| Code migration (Vals AI)Coding | 19.9% | Not in index |
| ProgramBench (Vals AI)Coding | Not measured | 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 | 53.6% | Not in index |
| SWE-bench Multimodal (Vals AI)Coding | 33.3% | Not in index |
| Vibe Code Bench 1-100 (Vals AI)Coding | 9.2% | Not in index |
| SkillsBench (Vals AI)Coding | 51.5% | Not in index |
| CyberBench (Vals AI)Coding | 65.0% | Not in index |
| Finance Agent (Vals AI)Agentic work | 48.3% | Not in index |
| Tax Agent Bench (Vals AI)Agentic work | 20.6% | Not in index |
| Legal Research Bench (Vals AI)Agentic work | 29.8% | Not in index |
| Harvey's Legal Agent Benchmark (Vals AI)Agentic work | Not measured | Not in index |
| Excel Modeling Benchmark (Vals AI)Agentic work | 47.8% | 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 | 64.1% | Not in index |
| MysteryMechanism (Vals AI)Reasoning | Not measured | Not in index |
| MMMU Pro (Vals AI)Reasoning | 81.2% | 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 | 50.6% | Not in index |
| LegalBench (Vals AI)Knowledge | 85.4% | Not in index |
| CaseLaw (Vals AI)Knowledge | Not measured | Not in index |
| CorpFin (Vals AI)Knowledge | 68.1% | Not in index |
| TaxEval (Vals AI)Knowledge | 72.7% | Not in index |
| MortgageTax (Vals AI)Knowledge | 68.4% | Not in index |
| MedQA (Vals AI)Knowledge | Not measured | Not in index |
| MedCode (Vals AI)Knowledge | 46.3% | Not in index |
| MedScribe (Vals AI)General | 87.3% | Not in index |
| Vals Multimodal IndexGeneral | 60.0% | Not in index |
| AA Math IndexMath | Not measured | Not in index |
| GPQA Diamond (Artificial Analysis run)Reasoning | 92.9% | Not in index |
| Humanity's Last Exam (Artificial Analysis run)Reasoning | 39.0% | 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 | 47.1% | Not in index |
| Terminal-Bench Hard (Artificial Analysis run)Coding | 42.4% | Not in index |
| IFBenchGeneral | 82.9% | Not in index |
| Long context reasoning (AA-LCR)General | 83.0% | Not in index |
| τ²-Bench TelecomAgentic work | 88.9% | Not in index |
| AIME 2025 (Artificial Analysis run)Math | Not measured | Not in index |
| Terminal-BenchCoding | Not measured | Not in index |
| DeepSWECoding | Not measured | Not in index |
| FrontierCodeCoding | 14.7% | Not in index |
| FrontierSWECoding | Not measured | Not in index |
| GSO-BenchCoding | Not measured | Not in index |
| WeirdMLCoding | Not measured | Not in index |
| MirrorCodeCoding | Not measured | Not in index |
| APEX-AgentsAgentic work | 37.7% | 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 | 4.6% | Not in index |
| BALROGAgentic work | Not measured | Not in index |
| DeepResearch BenchAgentic work | Not measured | Not in index |
| PostTrainBenchAgentic work | Not measured | Not in index |
| GPQA Diamond (Epoch AI run)Reasoning | 90.9% | Not in index |
| ARC-AGIReasoning | Not measured | Not in index |
| ARC-AGI-2Reasoning | Not measured | Not in index |
| Humanity's Last ExamReasoning | Not measured | Not in index |
| SimpleBenchReasoning | 45.8% | Not in index |
| Chess puzzlesReasoning | 13.0% | Not in index |
| Mystery game puzzlesReasoning | 8.0% | Not in index |
| EBR-benchReasoning | Not measured | Not in index |
| OTIS Mock AIMEMath | 71.1% | Not in index |
| FrontierMath, tiers 1 to 3Math | Not measured | Not in index |
| FrontierMath, tier 4Math | Not measured | Not in index |
| ProofBenchMath | 18.0% | Not in index |
| MATH level 5Math | Not measured | Not in index |
| GSM8KMath | Not measured | Not in index |
| SimpleQA VerifiedKnowledge | Not measured | Not in index |
| MMLUKnowledge | Not measured | Not in index |
| GeoBenchKnowledge | Not measured | Not in index |
| DTBenchGeneral | 78.9% | Not in index |
| LMCAGeneral | 33.7% | 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 | 55.0% | Not in index |
| Furniture assemblyGeneral | Not measured | Not in index |
| ExploitBenchGeneral | Not measured | 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 | 640.0 | Not in index |
| Vending-Bench 2Agentic work | $2,158 | Not in index |
| Epoch Capabilities IndexIntelligence | 147.1 | Not in index |
| LMArena TextGeneral | 1433 | Not in index |
| LMArena Text: codingCoding | 1471 | Not in index |
| LMArena Text: hard promptsReasoning | 1448 | Not in index |
| LMArena Text: mathMath | 1429 | Not in index |
| LMArena Text: instruction followingGeneral | 1433 | Not in index |
| LMArena Text: creative writingGeneral | 1404 | Not in index |
| LMArena VisionGeneral | 1251 | Not in index |
| LMArena Text: EnglishGeneral | 1444 | Not in index |
| LMArena Text: non-EnglishGeneral | 1420 | Not in index |
| LMArena Text: ChineseGeneral | 1467 | Not in index |
| LMArena Text: GermanGeneral | 1427 | Not in index |
| LMArena Text: JapaneseGeneral | 1381 | Not in index |
| LMArena Text: KoreanGeneral | 1373 | Not in index |
| LMArena Text: FrenchGeneral | 1445 | Not in index |
| LMArena Text: SpanishGeneral | 1432 | Not in index |
| LMArena Text: RussianGeneral | 1427 | Not in index |
| LMArena Text: PolishGeneral | 1424 | Not in index |
| LMArena Text: multi-turnGeneral | 1443 | Not in index |
| LMArena Text: longer queriesGeneral | 1446 | Not in index |
| LMArena Text: expertReasoning | 1463 | Not in index |
| LMArena Text: hard prompts in EnglishReasoning | 1456 | Not in index |
| LMArena Text: software and ITCoding | 1467 | Not in index |
| LMArena Text: mathematical workMath | 1447 | Not in index |
| LMArena Text: scienceKnowledge | 1453 | Not in index |
| LMArena Text: medicine and healthcareKnowledge | 1442 | Not in index |
| LMArena Text: legal and governmentKnowledge | 1438 | Not in index |
| LMArena Text: business and financeKnowledge | 1435 | Not in index |
| LMArena Text: writing and languageGeneral | 1420 | Not in index |
| LMArena Text: entertainment and mediaGeneral | 1403 | Not in index |
| LMArena WebDev: ReactCoding | 1478 | Not in index |
| LMArena WebDev: HTMLCoding | 1455 | Not in index |
| LMArena WebDev: from an imageCoding | 1470 | Not in index |
| LMArena Vision: EnglishGeneral | 1263 | Not in index |
| LMArena Vision: ChineseGeneral | 1311 | Not in index |
| LMArena Vision: reading textGeneral | 1259 | Not in index |
| LMArena Vision: diagramsReasoning | 1266 | Not in index |
| LMArena Vision: homeworkReasoning | 1268 | Not in index |
| LMArena Vision: humorGeneral | 1235 | Not in index |
| LMArena Vision: captioningGeneral | Not measured | Not in index |
| LMArena Vision: recognizing thingsGeneral | 1197 | Not in index |
| LMArena Vision: creative writingGeneral | 1234 | Not in index |
| LMArena SearchAgentic work | Not measured | Not in index |
| LMArena DocumentGeneral | 1435 | 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 | 70.9% | $0.4551 per task | Output tokens per task 38,007 |
| LiveBench reasoning | 74.5% | $0.0194 per task | Output tokens per task 15,833 |
| LiveBench data analysis | 76.2% | $0.0288 per task | Output tokens per task 23,291 |
| LiveBench research | 68.0% | $0.2516 per task | Output tokens per task 34,479 |
| LiveBench browser | 8.0% | $0.4292 per task | Output tokens per task 29,095 |
| LiveBench long agentic | 12.4% | $4.1953 per task | Output tokens per task 254,681 |
| LiveBench design | 66.2% | $0.2079 per task | Output tokens per task 46,783 |
| SWE-bench Verified (Vals AI) | 75.0% | $0.42 per test | Duration, as printed 12m 06s |
| Terminal-Bench 4 (Vals AI) | 1.0% | $7.23 per test | Duration, as printed 1h 16m |
| Vibe Code Bench (Vals AI) | 47.6% | $6.45 per test | Duration, as printed 1h 25m |
| Code migration (Vals AI) | 19.9% | $7.07 per test | Duration, as printed 1h 14m |
| Terminal-Bench 2.1 (Vals AI) | 53.6% | $0.21 per test | Duration, as printed 17m 03s |
| SWE-bench Multimodal (Vals AI) | 33.3% | $1.02 per test | Duration, as printed 13m 09s |
|---|---|---|---|
| Vibe Code Bench 1-100 (Vals AI) | 9.2% | $3.03 per test | Duration, as printed 26m 10s |
| SkillsBench (Vals AI) | 51.5% | $0.50 per test | Duration, as printed 15m 28s |
| CyberBench (Vals AI) | 65.0% | $3.41 per test | Duration, as printed 40m 56s |
| Finance Agent (Vals AI) | 48.3% | $0.32 per test | Duration, as printed 8m 17s |
| Tax Agent Bench (Vals AI) | 20.6% | $0.16 per test | Duration, as printed 5m 15s |
| Legal Research Bench (Vals AI) | 29.8% | $0.34 per test | Duration, as printed 13m 34s |
| Excel Modeling Benchmark (Vals AI) | 47.8% | $2.09 per test | Duration, as printed 31m 30s |
| Public Benefits Bench (Vals AI) | 64.1% | $0.25 per test | Nothing else |
| Vals Multimodal Index | 60.0% | $1.08 per test | Duration, as printed 17m 56s |
| OSWorld 2.0 | 4.6% | $258.78 estimated cost | With partial credit 22.3% |
| ProofBench | 18.0% | $0.4192 per test | Time per proof 11m 59s |
| Surface Evolver Bench | 55.0% | $2.0781 of the run | Pass rate 18.8% · Tokens for the run 5.5M |
| ALE-Bench | 640.0 | $0.1118 per problem | Tokens per problem 55,120 |
| LMArena WebDev | 1482 Elo | Not published | Votes 19,803 · Interval (±) ±5.5 |
| GPQA Diamond | 92.7% | Not published | Latency, as printed 4m 39s |
| MMLU Pro | 84.2% | Not published | Latency, as printed 41 s |
| Design Arena: websites | 1258 Elo | Not published | Battles 16,301 |
| Design Arena: UI components | 1247 Elo | Not published | Battles 1,841 |
| Design Arena: game dev | 1210 Elo | Not published | Battles 4,423 |
| Design Arena: data viz | 1235 Elo | Not published | Battles 1,387 |
| Design Arena: 3D | 1209 Elo | Not published | Battles 4,883 |
| Design Arena: all code categories | 1247 Elo | Not published | Battles 27,491 |
| Design Arena: SVG | 1161 Elo | Not published | Battles 2,593 |
| Design Arena: ASCII art | 1161 Elo | Not published | Battles 500 |
| Design Arena: agents building web apps | 1198 Elo | Not published | Battles 3,567 |
| Design Arena: agents building full-stack apps | 1197 Elo | Not published | Battles 3,673 |
| Design Arena: agents building mobile apps | 1168 Elo | Not published | Battles 2,154 |
| Design Arena: agents building Android apps | 1173 Elo | Not published | Battles 1,160 |
| Design Arena: agents building games | 1156 Elo | Not published | Battles 514 |
| Design Arena: HTML slides | 1180 Elo | Not published | Battles 1,246 |
| Design Arena: PowerPoint slides | 1205 Elo | Not published | Battles 1,058 |
| Design Arena: spreadsheets | 1250 Elo | Not published | Battles 35 |
| LiveCodeBench (Vals AI) | 82.2% | Not published | Latency, as printed 6m 07s |
| MMMU Pro (Vals AI) | 81.2% | Not published | Latency, as printed 1m 08s |
| SAGE (Vals AI) | 50.6% | Not published | Latency, as printed 1m 59s |
| LegalBench (Vals AI) | 85.4% | Not published | Latency, as printed 8.2 s |
| CorpFin (Vals AI) | 68.1% | Not published | Latency, as printed 24 s |
| TaxEval (Vals AI) | 72.7% | Not published | Latency, as printed 1m 34s |
| MortgageTax (Vals AI) | 68.4% | Not published | Latency, as printed 27 s |
| MedCode (Vals AI) | 46.3% | Not published | Latency, as printed 1m 03s |
| MedScribe (Vals AI) | 87.3% | Not published | Latency, as printed 2m 04s |
| APEX-Agents | 37.7% | Not published | Mean score 53.5% |
| LMArena Text | 1433 Elo | Not published | Votes 63,973 · Interval (±) ±3.8 |
| LMArena Text: coding | 1471 Elo | Not published | Votes 18,242 · Interval (±) ±5.6 |
| LMArena Text: hard prompts | 1448 Elo | Not published | Votes 42,422 · Interval (±) ±4.5 |
| LMArena Text: math | 1429 Elo | Not published | Votes 3,192 · Interval (±) ±10.9 |
| LMArena Text: instruction following | 1433 Elo | Not published | Votes 22,259 · Interval (±) ±5.3 |
| LMArena Text: creative writing | 1404 Elo | Not published | Votes 11,713 · Interval (±) ±6.6 |
| LMArena Vision | 1251 Elo | Not published | Votes 19,844 · Interval (±) ±6.3 |
| LMArena Text: English | 1444 Elo | Not published | Votes 26,416 · Interval (±) ±5.0 |
| LMArena Text: non-English | 1420 Elo | Not published | Votes 37,548 · Interval (±) ±4.5 |
| LMArena Text: Chinese | 1467 Elo | Not published | Votes 4,069 · Interval (±) ±10.0 |
| LMArena Text: German | 1427 Elo | Not published | Votes 1,104 · Interval (±) ±18.8 |
| LMArena Text: Japanese | 1381 Elo | Not published | Votes 953 · Interval (±) ±21.4 |
| LMArena Text: Korean | 1373 Elo | Not published | Votes 1,192 · Interval (±) ±18.6 |
| LMArena Text: French | 1445 Elo | Not published | Votes 2,295 · Interval (±) ±14.2 |
| LMArena Text: Spanish | 1432 Elo | Not published | Votes 1,959 · Interval (±) ±14.6 |
| LMArena Text: Russian | 1427 Elo | Not published | Votes 7,506 · Interval (±) ±7.5 |
| LMArena Text: Polish | 1424 Elo | Not published | Votes 1,112 · Interval (±) ±17.8 |
| LMArena Text: multi-turn | 1443 Elo | Not published | Votes 11,126 · Interval (±) ±6.6 |
| LMArena Text: longer queries | 1446 Elo | Not published | Votes 28,906 · Interval (±) ±5.1 |
| LMArena Text: expert | 1463 Elo | Not published | Votes 6,992 · Interval (±) ±7.8 |
| LMArena Text: hard prompts in English | 1456 Elo | Not published | Votes 18,062 · Interval (±) ±5.6 |
| LMArena Text: software and IT | 1467 Elo | Not published | Votes 25,769 · Interval (±) ±5.0 |
| LMArena Text: mathematical work | 1447 Elo | Not published | Votes 3,630 · Interval (±) ±10.5 |
| LMArena Text: science | 1453 Elo | Not published | Votes 10,165 · Interval (±) ±6.6 |
| LMArena Text: medicine and healthcare | 1442 Elo | Not published | Votes 4,321 · Interval (±) ±9.7 |
| LMArena Text: legal and government | 1438 Elo | Not published | Votes 5,214 · Interval (±) ±8.9 |
| LMArena Text: business and finance | 1435 Elo | Not published | Votes 12,339 · Interval (±) ±6.3 |
| LMArena Text: writing and language | 1420 Elo | Not published | Votes 16,321 · Interval (±) ±5.8 |
| LMArena Text: entertainment and media | 1403 Elo | Not published | Votes 15,034 · Interval (±) ±6.0 |
| LMArena WebDev: React | 1478 Elo | Not published | Votes 13,888 · Interval (±) ±6.2 |
| LMArena WebDev: HTML | 1455 Elo | Not published | Votes 2,689 · Interval (±) ±12.6 |
| LMArena WebDev: from an image | 1470 Elo | Not published | Votes 5,983 · Interval (±) ±8.5 |
| LMArena Vision: English | 1263 Elo | Not published | Votes 7,372 · Interval (±) ±9.0 |
| LMArena Vision: Chinese | 1311 Elo | Not published | Votes 1,048 · Interval (±) ±20.2 |
| LMArena Vision: reading text | 1259 Elo | Not published | Votes 13,947 · Interval (±) ±6.5 |
| LMArena Vision: diagrams | 1266 Elo | Not published | Votes 5,289 · Interval (±) ±9.2 |
| LMArena Vision: homework | 1268 Elo | Not published | Votes 2,405 · Interval (±) ±12.7 |
| LMArena Vision: humor | 1235 Elo | Not published | Votes 773 · Interval (±) ±22.1 |
| LMArena Vision: recognizing things | 1197 Elo | Not published | Votes 107 · Interval (±) ±53.1 |
| LMArena Vision: creative writing | 1234 Elo | Not published | Votes 1,339 · Interval (±) ±17.0 |
| LMArena Document | 1435 Elo | Not published | Votes 6,314 · Interval (±) ±8.3 |