🆕 NewJuly 15, 2026
Inkling released by Thinking Machines
Mira Murati's Thinking Machines Lab ships its first model: a 975B open-weights multimodal flagship under Apache 2.0, available on Hugging Face.
Inkling
🧠 Reasoning
Thinking Machines' first model: a 975B-parameter open-weights multimodal flagship under Apache 2.0. The leading US open model on real-world coding.
Released July 15, 2026
Pricing
Input tokensN/A/M
Output tokensN/A/M
Capacity
Context window1.0M tokens
Capabilities
✓Reasoning & Planning
✗Web Search
✓Open Source
Best Scores
GDPval-AA v21237.0 pts
GPQA Diamond87.2%
SWE-bench Verified77.6%
Terminal-Bench 2.163.8%
SWE-bench Pro54.3%
Best For
💻coding
🔬research
📊analysis
Benchmark Scores
Specialized Skills
GDPval-AA v2
Real paid work from 44 different jobs, such as law, nursing, and software. Judges compare two answers side by side without knowing which model wrote them, and the winner gains rating points. A typical human expert scores 1000, so a higher number means the work was picked over a human more often.
Independently measured by Artificial Analysis.
Knowledge
GPQA Diamond
PhD-level science questions written so you cannot just Google the answer. Tests whether the model can reason about hard science.
Provider-reported; no independent run recorded yet.
Humanity's Last Exam
PhD-level questions across many subjects. Tests deep reasoning on the hardest questions humans can ask.
Provider-reported; no independent run recorded yet.
Software Engineering
SWE-bench Verified
Hands the AI bugs from actual software projects and counts how many it fixes. Like a coding job interview, but with real work.
Provider-reported; no independent run recorded yet.
Terminal-Bench 2.1
Puts the AI in front of a computer terminal and asks it to finish multi-step tasks on its own. Measures how good an "AI agent" it is.
Provider-reported; an independent run by Vals AI (Terminus 2) lands at 47.57%.
SWE-bench Pro
The harder version of the coding test. Bigger codebases, trickier bugs. Scores drop for everyone, so the gaps between models become clearer.
Provider-reported; no independent run recorded yet.
Why Choose Inkling?
Inkling is the strongest US-built open-weights model, and the first flagship from Mira Murati's Thinking Machines Lab. It handles text, images, and audio natively, and the Apache 2.0 license is as business-friendly as it gets. Weights are on Hugging Face, with hosted APIs available from several providers.
How It Compares
vs GLM-5.2
Inkling leads on SWE-bench coding and is natively multimodal; GLM-5.2 is stronger on science reasoning and terminal work.
vs DeepSeek V4 Pro
Inkling adds vision and audio; DeepSeek edges it on pure coding benchmarks.