Private AI
The non-thinking version of Thinking Machines' 975B-parameter open-weights Mixture-of-Experts generalist with 41B active parameters. It gives faster direct answers across text, images, and audio, and is built for agentic coding, tool use, detailed instruction following, and long-context work.
Added Jul 15, 2026
Model weightsContext Window
1.0M
Max Output
32.8K
Input Price (Auto)
$1.05/1M
Output Price (Auto)
$4.25/1M
Cache Read (Auto)
$0.18/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
40.7
Choose explicit providers for this model. Auto routing remains available as the default option.
Loading provider options…
Coding Index
52.1
Agentic Index
32.3
GPQA Diamond
Graduate-level scientific reasoning
87.2%
Better than 87% of models compared
HLE
Humanity's Last Exam
29.7%
Better than 88% of models compared
AA-LCR
Long context reasoning evaluation
63.3%
Better than 76% of models compared
GDPval-AA
Economically valuable tasks
36.9%
CritPt
Research-level physics reasoning
5.4%
SciCode
Python programming for scientific computing
46.1%
Better than 86% of models compared
AA-Omniscience Accuracy
Proportion of correctly answered questions
40.0%
AA-Omniscience Hallucination Rate
Rate of incorrect answers among non-correct responses
63.1%
Last updated Aug 5, 2026
Artificial Analysis