#23Thinking Machines· released 2026-07-17

Thinking Machines: Inkling.

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,...

View on OpenRouter ↗
Context
1.0M
Input $/1M
$1.00
Output $/1M
$4.05
Modality
text · image · audio
Arena Elo
1543
LMArena — human blind votes
AA Index
60.0
Artificial Analysis composite
LiveBench
79.1
Contamination-resistant
GPQA
81.6%
Grad-level science reasoning
MMLU-Pro
81.8%
Broad knowledge
SWE-bench
67.8%
Real GitHub bugs (Verified)
Terminal-Bench
54.1%
Agentic shell tasks
ARC-AGI
14.9%
Abstract reasoning
Aider Polyglot
75.9%
Multi-language code editing
Speed
67 tok/s
Median output speed

Profile

Thinking Machines: Inkling is Thinking Machines's entry at position #23 on the current aggregated leaderboard. Its strongest signals are graduate-level science reasoning, real-world software engineering, unattended agentic work, and very long context. Every figure below is aggregated from the public sources listed on the homepage and refreshed on each pipeline run.

The score spread is comparatively complete for this model, but the standing advice still applies: treat gaps of a few points as ties and validate on your own workload before committing.

How this model's position is computed is documented in the methodology; what each benchmark actually measures is in the benchmark guide. See a figure that disagrees with its source? Report it — accepted corrections apply on the next refresh.

Nearby on the leaderboard