Translate
Independent model intelligence

Find the right AI model,
not just the loudest one.

Compare current model capability, context, modalities, public preference and API economics in one continuously refreshed decision surface.

LiveBench 2.0 LMArena models.dev OpenRouter AIHOT cross-check
Back to leaderboard
Thinking Machines · Active

Inkling

ling family · launched 2026-07-15

MultimodalReasoningToolsOpen weights
AITopic score60.5#56 overall

Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and tool-use systems,...

Evidence coverage95%5 evaluations · 2 institutions
Context / maximum output524.3k / 471.9ktokens
Cache / input / output$0.17 · $1 · $4.05USD per million tokens
Value / freshness56.8 / 76relative value · release freshness
Product profile

What this model supports

Inputtext, image, audio
Outputtext
Knowledge cutoffNot disclosed
WeightsOpen weights
Capability profile

Strengths by domain

Coding64.01 public results
Research & reasoning78.31 public results
Professional work37.91 public results
Knowledge & instruction59.91 public results
Behind the score

Every result has a source

Raw results and standardized readings are shown together. Missing tests add neither points nor penalties.

Coding1 results

LiveBench 2.0 · CodingLiveBench64.00 scoreReading 64.0

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0Open evidence

Research & reasoning1 results

LiveBench 2.0 · ResearchLiveBench78.30 scoreReading 78.3

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0Open evidence

Professional work1 results

LiveBench 2.0 · ProfessionalLiveBench37.90 scoreReading 37.9

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized. Professional = long-agentic, browser, data-analysis and design average.

Configuration: LiveBench 2.0Open evidence

Knowledge & instruction1 results

LiveBench 2.0 · KnowledgeLiveBench59.90 scoreReading 59.9

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized. Knowledge uses the published LiveBench research suite.

Configuration: LiveBench 2.0Open evidence

Human preference & indices3 results

Arena TextLMArena1,442.7 EloReading 60.5

Direct LMArena text leaderboard result.

Open evidence
AA IndexArtificial Analysis63.99 indexReading 55.3

Imported from Artificial Analysis. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0Open evidence
LiveBenchLiveBench50.54 scoreReading 50.5

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0Open evidence
Same builder

More from Thinking Machines