As of October 20, 2026, Google Gemini 3.5 Flash ranks #45 of 213 models overall with 33.6 on the Artificial Analysis Intelligence Index. Its strongest category is knowledge, where it ranks #9 of 147. Google Gemini 3.5 Flash costs $1.50 per million input tokens and $9.00 per million output tokens. It was released on May 19, 2026.
Its output price ranks 281 of 349 priced models, counting from the cheapest.
Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution...
google/gemini-3.5-flash 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 5 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, AA Coding Index, IOI (Vals AI), SWE-bench Multimodal (Vals AI), Vibe Code Bench 1-100 (Vals AI), CyberBench (Vals AI), LiveCodeBench (Artificial Analysis run), Terminal-Bench, FrontierCode, FrontierSWE, GSO-Bench, MirrorCode, Cybench
Not measured on AA Agentic Index, Design Arena: spreadsheets, LiveBench research, LiveBench browser, LiveBench long agentic, SREBench (Vals AI), Harvey's Legal Agent Benchmark (Vals AI), BioMysteryBench (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, OSWorld 2.0, BALROG, DeepResearch Bench, PostTrainBench, TheAgentCompany, OSWorld, SpireClimb, LMArena Search
Not measured on LiveBench reasoning, MysteryMechanism (Vals AI), Humanity's Last Exam, VPCT, BIG-Bench Hard
Not measured on AIME, MATH 500 (Vals AI), MGSM (Vals AI), AA Math Index, AIME 2025 (Artificial Analysis run), MATH level 5, GSM8K
Not measured on CaseLaw (Vals AI), MedQA (Vals AI), MMLU Pro (Artificial Analysis run), MMLU, GeoBench, AI2 Reasoning Challenge
Not measured on LiveBench design
Not measured on Vals Index, LiveBench data analysis, Fiction.LiveBench, Creative writing (Lech Mazur), CL-bench, Furniture assembly, ExploitBench, HellaSwag, WinoGrande, PIQA, CL-bench Life, LMArena Vision: captioning, LMArena Vision: recognizing things
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 | Not measured | 20% |
| LMArena WebDevCoding | 1499 | 15% |
| Vals IndexGeneral | Not measured | Not in index |
| GPQA DiamondReasoning | 92.7% | Not in index |
| AIMEMath | Not measured | Not in index |
| MMLU ProKnowledge | 89.5% | Not in index |
| AA Intelligence IndexIntelligence | 33.6 | Not in index |
| AA Coding IndexCoding | Not measured | Not in index |
| AA Agentic IndexAgentic work | Not measured | Not in index |
| Design Arena: websitesWeb design | 1268 | Not in index |
| Design Arena: UI componentsWeb design | 1273 | Not in index |
| Design Arena: game devWeb design | 1263 | Not in index |
| Design Arena: data vizWeb design | 1236 | Not in index |
| Design Arena: 3DWeb design | 1244 | Not in index |
| Design Arena: all code categoriesWeb design | 1266 | Not in index |
| Design Arena: SVGWeb design | 1253 | Not in index |
| Design Arena: ASCII artWeb design | 1254 | Not in index |
| Design Arena: agents building web appsAgentic work | 1202 | Not in index |
| Design Arena: agents building full-stack appsAgentic work | 1197 | Not in index |
| Design Arena: agents building mobile appsAgentic work | 1176 | Not in index |
| Design Arena: agents building Android appsAgentic work | 1200 | Not in index |
| Design Arena: agents building Godot gamesAgentic work | 1098 | Not in index |
| Design Arena: agents building gamesAgentic work | 1180 | Not in index |
| Design Arena: HTML slidesAgentic work | 1142 | Not in index |
| Design Arena: PowerPoint slidesAgentic work | 1178 | 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 | Not measured | Not in index |
| LiveBench data analysisGeneral | Not measured | Not in index |
| LiveBench researchAgentic work | Not measured | Not in index |
| LiveBench browserAgentic work | Not measured | Not in index |
| LiveBench long agenticAgentic work | Not measured | Not in index |
| LiveBench designWeb design | Not measured | Not in index |
| LiveCodeBench (Vals AI)Coding | 87.6% | Not in index |
| SWE-bench Verified (Vals AI)Coding | 78.8% | Not in index |
| Terminal-Bench 4 (Vals AI)Coding | 6.1% | Not in index |
| Vibe Code Bench (Vals AI)Coding | 48.7% | Not in index |
| Code migration (Vals AI)Coding | 26.8% | 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 | 74.2% | 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 | 52.7% | Not in index |
| CyberBench (Vals AI)Coding | Not measured | Not in index |
| Finance Agent (Vals AI)Agentic work | 57.9% | Not in index |
| Tax Agent Bench (Vals AI)Agentic work | 21.6% | Not in index |
| Legal Research Bench (Vals AI)Agentic work | 30.8% | Not in index |
| Harvey's Legal Agent Benchmark (Vals AI)Agentic work | Not measured | Not in index |
| Excel Modeling Benchmark (Vals AI)Agentic work | 63.5% | Not in index |
| BioMysteryBench (Vals AI)Agentic work | Not measured | Not in index |
| Terminal-Bench-Science (Vals AI)Agentic work | 5.7% | 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 | 59.5% | Not in index |
| MysteryMechanism (Vals AI)Reasoning | Not measured | Not in index |
| MMMU Pro (Vals AI)Reasoning | 88.3% | 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 | 49.9% | Not in index |
| LegalBench (Vals AI)Knowledge | 83.6% | Not in index |
| CaseLaw (Vals AI)Knowledge | Not measured | Not in index |
| CorpFin (Vals AI)Knowledge | 64.7% | Not in index |
| TaxEval (Vals AI)Knowledge | 74.4% | Not in index |
| MortgageTax (Vals AI)Knowledge | 68.1% | Not in index |
| MedQA (Vals AI)Knowledge | Not measured | Not in index |
| MedCode (Vals AI)Knowledge | 55.8% | Not in index |
| MedScribe (Vals AI)General | 76.6% | Not in index |
| Vals Multimodal IndexGeneral | 62.9% | Not in index |
| AA Math IndexMath | Not measured | Not in index |
| GPQA Diamond (Artificial Analysis run)Reasoning | 92.2% | Not in index |
| Humanity's Last Exam (Artificial Analysis run)Reasoning | 42.7% | 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 | 53.9% | Not in index |
| Terminal-Bench Hard (Artificial Analysis run)Coding | 46.2% | Not in index |
| IFBenchGeneral | 76.3% | Not in index |
| Long context reasoning (AA-LCR)General | 74.3% | Not in index |
| τ²-Bench TelecomAgentic work | 95.6% | Not in index |
| AIME 2025 (Artificial Analysis run)Math | Not measured | Not in index |
| Terminal-BenchCoding | Not measured | Not in index |
| DeepSWECoding | 37.4% | Not in index |
| FrontierCodeCoding | Not measured | Not in index |
| FrontierSWECoding | Not measured | Not in index |
| GSO-BenchCoding | Not measured | Not in index |
| WeirdMLCoding | 62.6% | Not in index |
| MirrorCodeCoding | Not measured | Not in index |
| APEX-AgentsAgentic work | 27.5% | 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 | 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 | 92.8% | Not in index |
| ARC-AGIReasoning | 92.5% | Not in index |
| ARC-AGI-2Reasoning | 72.1% | Not in index |
| Humanity's Last ExamReasoning | Not measured | Not in index |
| SimpleBenchReasoning | 76.7% | Not in index |
| Chess puzzlesReasoning | 45.0% | Not in index |
| Mystery game puzzlesReasoning | 32.0% | Not in index |
| EBR-benchReasoning | 4.8% | Not in index |
| OTIS Mock AIMEMath | 95.6% | Not in index |
| FrontierMath, tiers 1 to 3Math | 62.8% | Not in index |
| FrontierMath, tier 4Math | 26.8% | Not in index |
| ProofBenchMath | 31.0% | Not in index |
| MATH level 5Math | Not measured | Not in index |
| GSM8KMath | Not measured | Not in index |
| SimpleQA VerifiedKnowledge | 66.2% | Not in index |
| MMLUKnowledge | Not measured | Not in index |
| GeoBenchKnowledge | Not measured | Not in index |
| DTBenchGeneral | 94.7% | Not in index |
| LMCAGeneral | 47.1% | 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 | 58.1% | 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 | 911.0 | Not in index |
| Vending-Bench 2Agentic work | $5,396 | Not in index |
| Epoch Capabilities IndexIntelligence | 154.4 | Not in index |
| LMArena TextGeneral | 1482 | Not in index |
| LMArena Text: codingCoding | 1493 | Not in index |
| LMArena Text: hard promptsReasoning | 1488 | Not in index |
| LMArena Text: mathMath | 1503 | Not in index |
| LMArena Text: instruction followingGeneral | 1467 | Not in index |
| LMArena Text: creative writingGeneral | 1470 | Not in index |
| LMArena VisionGeneral | 1308 | Not in index |
| LMArena Text: EnglishGeneral | 1481 | Not in index |
| LMArena Text: non-EnglishGeneral | 1475 | Not in index |
| LMArena Text: ChineseGeneral | 1526 | Not in index |
| LMArena Text: GermanGeneral | 1493 | Not in index |
| LMArena Text: JapaneseGeneral | 1486 | Not in index |
| LMArena Text: KoreanGeneral | 1449 | Not in index |
| LMArena Text: FrenchGeneral | 1490 | Not in index |
| LMArena Text: SpanishGeneral | 1480 | Not in index |
| LMArena Text: RussianGeneral | 1492 | Not in index |
| LMArena Text: PolishGeneral | 1517 | Not in index |
| LMArena Text: multi-turnGeneral | 1481 | Not in index |
| LMArena Text: longer queriesGeneral | 1483 | Not in index |
| LMArena Text: expertReasoning | 1495 | Not in index |
| LMArena Text: hard prompts in EnglishReasoning | 1482 | Not in index |
| LMArena Text: software and ITCoding | 1494 | Not in index |
| LMArena Text: mathematical workMath | 1494 | Not in index |
| LMArena Text: scienceKnowledge | 1494 | Not in index |
| LMArena Text: medicine and healthcareKnowledge | 1483 | Not in index |
| LMArena Text: legal and governmentKnowledge | 1488 | Not in index |
| LMArena Text: business and financeKnowledge | 1467 | Not in index |
| LMArena Text: writing and languageGeneral | 1478 | Not in index |
| LMArena Text: entertainment and mediaGeneral | 1459 | Not in index |
| LMArena WebDev: ReactCoding | 1495 | Not in index |
| LMArena WebDev: HTMLCoding | 1522 | Not in index |
| LMArena WebDev: from an imageCoding | 1504 | Not in index |
| LMArena Vision: EnglishGeneral | 1305 | Not in index |
| LMArena Vision: ChineseGeneral | 1377 | Not in index |
| LMArena Vision: reading textGeneral | 1317 | Not in index |
| LMArena Vision: diagramsReasoning | 1323 | Not in index |
| LMArena Vision: homeworkReasoning | 1330 | Not in index |
| LMArena Vision: humorGeneral | 1321 | Not in index |
| LMArena Vision: captioningGeneral | Not measured | Not in index |
| LMArena Vision: recognizing thingsGeneral | Not measured | Not in index |
| LMArena Vision: creative writingGeneral | 1301 | Not in index |
| LMArena SearchAgentic work | Not measured | Not in index |
| LMArena DocumentGeneral | 1463 | 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 (Vals AI) | 78.8% | $0.95 per test | Duration, as printed 4m 14s |
| Terminal-Bench 4 (Vals AI) | 6.1% | $5.98 per test | Duration, as printed 1h 11m |
| Vibe Code Bench (Vals AI) | 48.7% | $2.54 per test | Duration, as printed 14m 53s |
| Code migration (Vals AI) | 26.8% | $5.73 per test | Duration, as printed 19m 28s |
| Terminal-Bench 2.1 (Vals AI) | 74.2% | $0.87 per test | Duration, as printed 6m 13s |
| SkillsBench (Vals AI) | 52.7% | $1.25 per test | Duration, as printed 7m 15s |
| Finance Agent (Vals AI) | 57.9% | $2.51 per test | Duration, as printed 5m 22s |
| Tax Agent Bench (Vals AI) | 21.6% | $0.86 per test | Duration, as printed 3m 53s |
| Legal Research Bench (Vals AI) | 30.8% | $1.31 per test | Duration, as printed 4m 38s |
| Excel Modeling Benchmark (Vals AI) | 63.5% | $5.17 per test | Duration, as printed 18m 44s |
| Terminal-Bench-Science (Vals AI) | 5.7% | $3.80 per test | Duration, as printed 1h 01m |
| Public Benefits Bench (Vals AI) | 59.5% | $1.11 per test | Nothing else |
| Vals Multimodal Index | 62.9% | $1.01 per test | Duration, as printed 4m 26s |
|---|---|---|---|
| DeepSWE | 37.4% | $7.3418 mean cost per task | Solved in 4 tries 66.4% · Output tokens per task 275,778 · Agent steps per task 86 |
| WeirdML | 62.6% | $0.756 per run | Median lines of code 286 |
| ARC-AGI | 92.5% | $0.4278 per task | Nothing else |
| ARC-AGI-2 | 72.1% | $0.8503 per task | Nothing else |
| ProofBench | 31.0% | $0.5699 per test | Time per proof 6m 34s |
| Surface Evolver Bench | 58.1% | $6.6382 of the run | Pass rate 37.5% · Tokens for the run 3.1M |
| ALE-Bench | 911.0 | $0.4419 per problem | Tokens per problem 51,880 |
| LMArena WebDev | 1499 Elo | Not published | Votes 9,382 · Interval (±) ±7.3 |
| GPQA Diamond | 92.7% | Not published | Latency, as printed 28 s |
| MMLU Pro | 89.5% | Not published | Latency, as printed 26 s |
| Design Arena: websites | 1268 Elo | Not published | Battles 20,793 |
| Design Arena: UI components | 1273 Elo | Not published | Battles 3,192 |
| Design Arena: game dev | 1263 Elo | Not published | Battles 6,101 |
| Design Arena: data viz | 1236 Elo | Not published | Battles 2,023 |
| Design Arena: 3D | 1244 Elo | Not published | Battles 7,356 |
| Design Arena: all code categories | 1266 Elo | Not published | Battles 36,840 |
| Design Arena: SVG | 1253 Elo | Not published | Battles 3,389 |
| Design Arena: ASCII art | 1254 Elo | Not published | Battles 1,384 |
| Design Arena: agents building web apps | 1202 Elo | Not published | Battles 5,699 |
| Design Arena: agents building full-stack apps | 1197 Elo | Not published | Battles 4,032 |
| Design Arena: agents building mobile apps | 1176 Elo | Not published | Battles 1,344 |
| Design Arena: agents building Android apps | 1200 Elo | Not published | Battles 2,407 |
| Design Arena: agents building Godot games | 1098 Elo | Not published | Battles 1,345 |
| Design Arena: agents building games | 1180 Elo | Not published | Battles 3,583 |
| Design Arena: HTML slides | 1142 Elo | Not published | Battles 2,196 |
| Design Arena: PowerPoint slides | 1178 Elo | Not published | Battles 4,040 |
| LiveCodeBench (Vals AI) | 87.6% | Not published | Latency, as printed 1m 04s |
| ProgramBench (Vals AI) | 0.0% | Not published | Duration, as printed 18m 25s |
| MMMU Pro (Vals AI) | 88.3% | Not published | Latency, as printed 12 s |
| SAGE (Vals AI) | 49.9% | Not published | Latency, as printed 33 s |
| LegalBench (Vals AI) | 83.6% | Not published | Latency, as printed 35 s |
| CorpFin (Vals AI) | 64.7% | Not published | Latency, as printed 8.7 s |
| TaxEval (Vals AI) | 74.4% | Not published | Latency, as printed 20 s |
| MortgageTax (Vals AI) | 68.1% | Not published | Latency, as printed 11 s |
| MedCode (Vals AI) | 55.8% | Not published | Latency, as printed 25 s |
| MedScribe (Vals AI) | 76.6% | Not published | Latency, as printed 57 s |
| APEX-Agents | 27.5% | Not published | Mean score 44.3% |
| LMArena Text | 1482 Elo | Not published | Votes 48,486 · Interval (±) ±3.9 |
| LMArena Text: coding | 1493 Elo | Not published | Votes 13,847 · Interval (±) ±6.0 |
| LMArena Text: hard prompts | 1488 Elo | Not published | Votes 32,403 · Interval (±) ±4.5 |
| LMArena Text: math | 1503 Elo | Not published | Votes 2,385 · Interval (±) ±12.4 |
| LMArena Text: instruction following | 1467 Elo | Not published | Votes 17,406 · Interval (±) ±5.6 |
| LMArena Text: creative writing | 1470 Elo | Not published | Votes 9,481 · Interval (±) ±7.3 |
| LMArena Vision | 1308 Elo | Not published | Votes 13,862 · Interval (±) ±6.8 |
| LMArena Text: English | 1481 Elo | Not published | Votes 20,571 · Interval (±) ±5.2 |
| LMArena Text: non-English | 1475 Elo | Not published | Votes 27,913 · Interval (±) ±4.7 |
| LMArena Text: Chinese | 1526 Elo | Not published | Votes 3,441 · Interval (±) ±10.7 |
| LMArena Text: German | 1493 Elo | Not published | Votes 758 · Interval (±) ±22.5 |
| LMArena Text: Japanese | 1486 Elo | Not published | Votes 695 · Interval (±) ±24.1 |
| LMArena Text: Korean | 1449 Elo | Not published | Votes 872 · Interval (±) ±20.9 |
| LMArena Text: French | 1490 Elo | Not published | Votes 1,544 · Interval (±) ±16.5 |
| LMArena Text: Spanish | 1480 Elo | Not published | Votes 1,430 · Interval (±) ±16.6 |
| LMArena Text: Russian | 1492 Elo | Not published | Votes 5,474 · Interval (±) ±8.6 |
| LMArena Text: Polish | 1517 Elo | Not published | Votes 799 · Interval (±) ±21.3 |
| LMArena Text: multi-turn | 1481 Elo | Not published | Votes 8,090 · Interval (±) ±7.4 |
| LMArena Text: longer queries | 1483 Elo | Not published | Votes 22,808 · Interval (±) ±5.3 |
| LMArena Text: expert | 1495 Elo | Not published | Votes 5,582 · Interval (±) ±8.5 |
| LMArena Text: hard prompts in English | 1482 Elo | Not published | Votes 13,789 · Interval (±) ±6.0 |
| LMArena Text: software and IT | 1494 Elo | Not published | Votes 19,417 · Interval (±) ±5.3 |
| LMArena Text: mathematical work | 1494 Elo | Not published | Votes 2,700 · Interval (±) ±12.0 |
| LMArena Text: science | 1494 Elo | Not published | Votes 7,792 · Interval (±) ±7.4 |
| LMArena Text: medicine and healthcare | 1483 Elo | Not published | Votes 3,559 · Interval (±) ±10.6 |
| LMArena Text: legal and government | 1488 Elo | Not published | Votes 3,849 · Interval (±) ±10.2 |
| LMArena Text: business and finance | 1467 Elo | Not published | Votes 9,395 · Interval (±) ±6.8 |
| LMArena Text: writing and language | 1478 Elo | Not published | Votes 12,654 · Interval (±) ±6.3 |
| LMArena Text: entertainment and media | 1459 Elo | Not published | Votes 12,062 · Interval (±) ±6.6 |
| LMArena WebDev: React | 1495 Elo | Not published | Votes 6,781 · Interval (±) ±8.4 |
| LMArena WebDev: HTML | 1522 Elo | Not published | Votes 1,342 · Interval (±) ±17.1 |
| LMArena WebDev: from an image | 1504 Elo | Not published | Votes 2,865 · Interval (±) ±11.7 |
| LMArena Vision: English | 1305 Elo | Not published | Votes 5,040 · Interval (±) ±9.9 |
| LMArena Vision: Chinese | 1377 Elo | Not published | Votes 847 · Interval (±) ±22.3 |
| LMArena Vision: reading text | 1317 Elo | Not published | Votes 9,712 · Interval (±) ±7.1 |
| LMArena Vision: diagrams | 1323 Elo | Not published | Votes 3,618 · Interval (±) ±10.6 |
| LMArena Vision: homework | 1330 Elo | Not published | Votes 1,553 · Interval (±) ±15.3 |
| LMArena Vision: humor | 1321 Elo | Not published | Votes 577 · Interval (±) ±25.4 |
| LMArena Vision: creative writing | 1301 Elo | Not published | Votes 1,180 · Interval (±) ±18.3 |
| LMArena Document | 1463 Elo | Not published | Votes 6,500 · Interval (±) ±7.9 |