As of October 20, 2026, Thinking Machines Inkling Small ranks #67 of 213 models overall with 25.7 on the Artificial Analysis Intelligence Index. Its strongest category is knowledge, where it ranks #56 of 147. Thinking Machines Inkling Small costs $0.45 per million input tokens and $1.20 per million output tokens. It was released on July 30, 2026.
Its output price ranks 147 of 349 priced models, counting from the cheapest.
Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...
thinkingmachines/inkling-small 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, SWE-bench Multimodal (Vals AI), Vibe Code Bench 1-100 (Vals AI), LiveCodeBench (Artificial Analysis run), Terminal-Bench Hard (Artificial Analysis run), Terminal-Bench, DeepSWE, FrontierCode, FrontierSWE, GSO-Bench, WeirdML, MirrorCode, Cybench, ALE-Bench
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, LiveBench research, LiveBench browser, LiveBench long agentic, SREBench (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, BALROG, DeepResearch Bench, PostTrainBench, TheAgentCompany, OSWorld, SpireClimb, Vending-Bench 2, LMArena Search
Not measured on LiveBench reasoning, MysteryMechanism (Vals AI), MMMU Pro (Vals AI), Humanity's Last Exam, SimpleBench, Chess puzzles, 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), 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 Design Arena: websites, Design Arena: UI components, Design Arena: game dev, Design Arena: data viz, Design Arena: 3D, Design Arena: all code categories, Design Arena: SVG, Design Arena: ASCII art, LiveBench design
Not measured on LiveBench data analysis, IFBench, DTBench, LMCA, Fiction.LiveBench, Creative writing (Lech Mazur), CL-bench, Surface Evolver Bench, Furniture assembly, ExploitBench, HellaSwag, WinoGrande, PIQA, CL-bench Life, LMArena Vision: captioning, LMArena Vision: recognizing things, LMArena Document
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 | 1409 | 15% |
| Vals IndexGeneral | 27.6% | Not in index |
| GPQA DiamondReasoning | 83.6% | Not in index |
| AIMEMath | Not measured | Not in index |
| MMLU ProKnowledge | 85.6% | Not in index |
| AA Intelligence IndexIntelligence | 25.7 | Not in index |
| AA Coding IndexCoding | 52.9 | Not in index |
| AA Agentic IndexAgentic work | 23.5 | Not in index |
| Design Arena: websitesWeb design | Not measured | Not in index |
| Design Arena: UI componentsWeb design | Not measured | Not in index |
| Design Arena: game devWeb design | Not measured | Not in index |
| Design Arena: data vizWeb design | Not measured | Not in index |
| Design Arena: 3DWeb design | Not measured | Not in index |
| Design Arena: all code categoriesWeb design | Not measured | Not in index |
| Design Arena: SVGWeb design | Not measured | Not in index |
| Design Arena: ASCII artWeb design | Not measured | 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 | 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 | 85.9% | Not in index |
| SWE-bench Verified (Vals AI)Coding | 82.2% | Not in index |
| Terminal-Bench 4 (Vals AI)Coding | 0.0% | Not in index |
| Vibe Code Bench (Vals AI)Coding | 19.1% | Not in index |
| Code migration (Vals AI)Coding | 13.7% | 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 | 9.3% | Not in index |
| Terminal-Bench 2.1 (Vals AI)Coding | 55.1% | 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 | 33.6% | Not in index |
| CyberBench (Vals AI)Coding | 64.1% | Not in index |
| Finance Agent (Vals AI)Agentic work | 41.3% | Not in index |
| Tax Agent Bench (Vals AI)Agentic work | 15.4% | Not in index |
| Legal Research Bench (Vals AI)Agentic work | 25.5% | Not in index |
| Harvey's Legal Agent Benchmark (Vals AI)Agentic work | 2.9% | Not in index |
| Excel Modeling Benchmark (Vals AI)Agentic work | 31.8% | 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 | 59.4% | Not in index |
| MysteryMechanism (Vals AI)Reasoning | Not measured | Not in index |
| MMMU Pro (Vals AI)Reasoning | Not measured | 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 | 32.7% | Not in index |
| LegalBench (Vals AI)Knowledge | 83.0% | Not in index |
| CaseLaw (Vals AI)Knowledge | Not measured | Not in index |
| CorpFin (Vals AI)Knowledge | 69.6% | Not in index |
| TaxEval (Vals AI)Knowledge | 75.5% | Not in index |
| MortgageTax (Vals AI)Knowledge | 62.3% | Not in index |
| MedQA (Vals AI)Knowledge | Not measured | Not in index |
| MedCode (Vals AI)Knowledge | 37.9% | Not in index |
| MedScribe (Vals AI)General | 84.1% | Not in index |
| Vals Multimodal IndexGeneral | 50.0% | Not in index |
| AA Math IndexMath | Not measured | Not in index |
| GPQA Diamond (Artificial Analysis run)Reasoning | 89.5% | Not in index |
| Humanity's Last Exam (Artificial Analysis run)Reasoning | 33.3% | 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 | 49.7% | 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 | 75.7% | 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 | Not measured | 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 | Not measured | 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 | 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 | 88.5% | Not in index |
| ARC-AGIReasoning | 84.0% | Not in index |
| ARC-AGI-2Reasoning | 40.1% | Not in index |
| Humanity's Last ExamReasoning | Not measured | Not in index |
| SimpleBenchReasoning | Not measured | Not in index |
| Chess puzzlesReasoning | Not measured | Not in index |
| Mystery game puzzlesReasoning | 6.0% | Not in index |
| EBR-benchReasoning | Not measured | Not in index |
| OTIS Mock AIMEMath | 90.0% | Not in index |
| FrontierMath, tiers 1 to 3Math | 46.3% | Not in index |
| FrontierMath, tier 4Math | 17.1% | Not in index |
| ProofBenchMath | 6.0% | Not in index |
| MATH level 5Math | Not measured | Not in index |
| GSM8KMath | Not measured | Not in index |
| SimpleQA VerifiedKnowledge | 19.1% | Not in index |
| MMLUKnowledge | Not measured | Not in index |
| GeoBenchKnowledge | Not measured | Not in index |
| DTBenchGeneral | Not measured | Not in index |
| LMCAGeneral | Not measured | 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 | 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 | Not measured | Not in index |
| Vending-Bench 2Agentic work | Not measured | Not in index |
| Epoch Capabilities IndexIntelligence | 150.2 | Not in index |
| LMArena TextGeneral | 1413 | Not in index |
| LMArena Text: codingCoding | 1452 | Not in index |
| LMArena Text: hard promptsReasoning | 1423 | Not in index |
| LMArena Text: mathMath | 1457 | Not in index |
| LMArena Text: instruction followingGeneral | 1399 | Not in index |
| LMArena Text: creative writingGeneral | 1331 | Not in index |
| LMArena VisionGeneral | 1237 | Not in index |
| LMArena Text: EnglishGeneral | 1421 | Not in index |
| LMArena Text: non-EnglishGeneral | 1400 | Not in index |
| LMArena Text: ChineseGeneral | 1463 | Not in index |
| LMArena Text: GermanGeneral | 1407 | Not in index |
| LMArena Text: JapaneseGeneral | 1401 | Not in index |
| LMArena Text: KoreanGeneral | 1364 | Not in index |
| LMArena Text: FrenchGeneral | 1433 | Not in index |
| LMArena Text: SpanishGeneral | 1422 | Not in index |
| LMArena Text: RussianGeneral | 1391 | Not in index |
| LMArena Text: PolishGeneral | 1426 | Not in index |
| LMArena Text: multi-turnGeneral | 1414 | Not in index |
| LMArena Text: longer queriesGeneral | 1401 | Not in index |
| LMArena Text: expertReasoning | 1442 | Not in index |
| LMArena Text: hard prompts in EnglishReasoning | 1429 | Not in index |
| LMArena Text: software and ITCoding | 1447 | Not in index |
| LMArena Text: mathematical workMath | 1441 | Not in index |
| LMArena Text: scienceKnowledge | 1427 | Not in index |
| LMArena Text: medicine and healthcareKnowledge | 1411 | Not in index |
| LMArena Text: legal and governmentKnowledge | 1416 | Not in index |
| LMArena Text: business and financeKnowledge | 1413 | Not in index |
| LMArena Text: writing and languageGeneral | 1355 | Not in index |
| LMArena Text: entertainment and mediaGeneral | 1351 | Not in index |
| LMArena WebDev: ReactCoding | 1399 | Not in index |
| LMArena WebDev: HTMLCoding | 1398 | Not in index |
| LMArena WebDev: from an imageCoding | 1410 | Not in index |
| LMArena Vision: EnglishGeneral | 1243 | Not in index |
| LMArena Vision: ChineseGeneral | 1292 | Not in index |
| LMArena Vision: reading textGeneral | 1243 | Not in index |
| LMArena Vision: diagramsReasoning | 1251 | Not in index |
| LMArena Vision: homeworkReasoning | 1257 | Not in index |
| LMArena Vision: humorGeneral | 1207 | Not in index |
| LMArena Vision: captioningGeneral | Not measured | Not in index |
| LMArena Vision: recognizing thingsGeneral | Not measured | Not in index |
| LMArena Vision: creative writingGeneral | 1205 | Not in index |
| LMArena SearchAgentic work | Not measured | Not in index |
| LMArena DocumentGeneral | Not measured | 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 |
|---|---|---|---|
| Vals Index | 27.6% | $0.40 per test | Duration, as printed 16m 26s |
| SWE-bench Verified (Vals AI) | 82.2% | $0.13 per test | Duration, as printed 4m 05s |
| Terminal-Bench 4 (Vals AI) | 0.0% | $0.32 per test | Duration, as printed 11m 45s |
| Vibe Code Bench (Vals AI) | 19.1% | $0.43 per test | Duration, as printed 49m 09s |
| Code migration (Vals AI) | 13.7% | $0.60 per test | Duration, as printed 20m 33s |
| ProgramBench (Vals AI) | 0.5% | $0.48 per test | Duration, as printed 28m 31s |
| IOI (Vals AI) | 9.3% | $0.34 per test | Duration, as printed 11m 18s |
| Terminal-Bench 2.1 (Vals AI) | 55.1% | $0.13 per test | Duration, as printed 10m 04s |
| SkillsBench (Vals AI) | 33.6% | $0.15 per test | Duration, as printed 11m 41s |
| CyberBench (Vals AI) | 64.1% | $0.25 per test | Duration, as printed 19m 35s |
| Finance Agent (Vals AI) | 41.3% | $0.23 per test | Duration, as printed 16m 34s |
| Tax Agent Bench (Vals AI) | 15.4% | $0.09 per test | Duration, as printed 6m 25s |
| Legal Research Bench (Vals AI) | 25.5% | $0.14 per test | Duration, as printed 12m 55s |
|---|---|---|---|
| Harvey's Legal Agent Benchmark (Vals AI) | 2.9% | $0.18 per test | Duration, as printed 7m 35s |
| Excel Modeling Benchmark (Vals AI) | 31.8% | $0.25 per test | Duration, as printed 19m 34s |
| Public Benefits Bench (Vals AI) | 59.4% | $0.10 per test | Nothing else |
| Vals Multimodal Index | 50.0% | $0.14 per test | Duration, as printed 11m 49s |
| ProofBench | 6.0% | $0.0754 per test | Time per proof 8m 27s |
| LMArena WebDev | 1409 Elo | Not published | Votes 5,323 · Interval (±) ±9.1 |
| GPQA Diamond | 83.6% | Not published | Latency, as printed 4m 20s |
| MMLU Pro | 85.6% | Not published | Latency, as printed 1m 12s |
| LiveCodeBench (Vals AI) | 85.9% | Not published | Latency, as printed 5m 02s |
| SAGE (Vals AI) | 32.7% | Not published | Latency, as printed 3m 26s |
| LegalBench (Vals AI) | 83.0% | Not published | Latency, as printed 18 s |
| CorpFin (Vals AI) | 69.6% | Not published | Latency, as printed 45 s |
| TaxEval (Vals AI) | 75.5% | Not published | Latency, as printed 4m 02s |
| MortgageTax (Vals AI) | 62.3% | Not published | Latency, as printed 1m 52s |
| MedCode (Vals AI) | 37.9% | Not published | Latency, as printed 3m 11s |
| MedScribe (Vals AI) | 84.1% | Not published | Latency, as printed 3m 50s |
| LMArena Text | 1413 Elo | Not published | Votes 32,304 · Interval (±) ±4.7 |
| LMArena Text: coding | 1452 Elo | Not published | Votes 8,848 · Interval (±) ±7.1 |
| LMArena Text: hard prompts | 1423 Elo | Not published | Votes 21,129 · Interval (±) ±5.4 |
| LMArena Text: math | 1457 Elo | Not published | Votes 1,455 · Interval (±) ±15.6 |
| LMArena Text: instruction following | 1399 Elo | Not published | Votes 11,768 · Interval (±) ±6.5 |
| LMArena Text: creative writing | 1331 Elo | Not published | Votes 6,963 · Interval (±) ±8.3 |
| LMArena Vision | 1237 Elo | Not published | Votes 8,074 · Interval (±) ±8.1 |
| LMArena Text: English | 1421 Elo | Not published | Votes 13,225 · Interval (±) ±6.1 |
| LMArena Text: non-English | 1400 Elo | Not published | Votes 19,075 · Interval (±) ±5.5 |
| LMArena Text: Chinese | 1463 Elo | Not published | Votes 2,589 · Interval (±) ±12.1 |
| LMArena Text: German | 1407 Elo | Not published | Votes 540 · Interval (±) ±25.9 |
| LMArena Text: Japanese | 1401 Elo | Not published | Votes 528 · Interval (±) ±27.9 |
| LMArena Text: Korean | 1364 Elo | Not published | Votes 583 · Interval (±) ±25.1 |
| LMArena Text: French | 1433 Elo | Not published | Votes 1,030 · Interval (±) ±19.7 |
| LMArena Text: Spanish | 1422 Elo | Not published | Votes 966 · Interval (±) ±19.5 |
| LMArena Text: Russian | 1391 Elo | Not published | Votes 3,866 · Interval (±) ±10.0 |
| LMArena Text: Polish | 1426 Elo | Not published | Votes 459 · Interval (±) ±27.9 |
| LMArena Text: multi-turn | 1414 Elo | Not published | Votes 4,945 · Interval (±) ±9.1 |
| LMArena Text: longer queries | 1401 Elo | Not published | Votes 15,204 · Interval (±) ±6.2 |
| LMArena Text: expert | 1442 Elo | Not published | Votes 3,781 · Interval (±) ±10.1 |
| LMArena Text: hard prompts in English | 1429 Elo | Not published | Votes 8,663 · Interval (±) ±7.2 |
| LMArena Text: software and IT | 1447 Elo | Not published | Votes 12,584 · Interval (±) ±6.2 |
| LMArena Text: mathematical work | 1441 Elo | Not published | Votes 1,748 · Interval (±) ±14.5 |
| LMArena Text: science | 1427 Elo | Not published | Votes 5,384 · Interval (±) ±8.6 |
| LMArena Text: medicine and healthcare | 1411 Elo | Not published | Votes 2,332 · Interval (±) ±12.9 |
| LMArena Text: legal and government | 1416 Elo | Not published | Votes 2,801 · Interval (±) ±11.8 |
| LMArena Text: business and finance | 1413 Elo | Not published | Votes 6,001 · Interval (±) ±8.3 |
| LMArena Text: writing and language | 1355 Elo | Not published | Votes 9,076 · Interval (±) ±7.3 |
| LMArena Text: entertainment and media | 1351 Elo | Not published | Votes 8,419 · Interval (±) ±7.6 |
| LMArena WebDev: React | 1399 Elo | Not published | Votes 3,554 · Interval (±) ±11.0 |
| LMArena WebDev: HTML | 1398 Elo | Not published | Votes 632 · Interval (±) ±25.7 |
| LMArena WebDev: from an image | 1410 Elo | Not published | Votes 1,025 · Interval (±) ±19.2 |
| LMArena Vision: English | 1243 Elo | Not published | Votes 2,816 · Interval (±) ±12.2 |
| LMArena Vision: Chinese | 1292 Elo | Not published | Votes 461 · Interval (±) ±28.4 |
| LMArena Vision: reading text | 1243 Elo | Not published | Votes 5,655 · Interval (±) ±8.6 |
| LMArena Vision: diagrams | 1251 Elo | Not published | Votes 2,071 · Interval (±) ±13.5 |
| LMArena Vision: homework | 1257 Elo | Not published | Votes 835 · Interval (±) ±20.4 |
| LMArena Vision: humor | 1207 Elo | Not published | Votes 384 · Interval (±) ±30.2 |
| LMArena Vision: creative writing | 1205 Elo | Not published | Votes 675 · Interval (±) ±24.0 |