From the source
- Thinking Machines released Inkling 2 Mini, an open mixture-of-experts model that runs certain layers repeatedly so it can reason internally for longer without writing more text.
- The company says the looped transformer runs 12 of its 24 layers three times in a row, letting it find answers with fewer tokens.
- It has 26.4 billion parameters, with 5.4 billion active per token because of the loops, and handles text and images with a 1-million-token context window under an Apache 2.0 license.
- Thinking Machines says it leads all open models under 100 billion parameters on agentic coding tests, scoring 48.2 percent on DeepSWE v1.1 and 46.5 percent on the Vals Index v2.2.
- Artificial Analysis gives it 20 points on its Intelligence Index, tying it with Qwen3.8 27B at roughly six times lower cost and four times the speed, according to the report.
- The model runs locally on Macs with at least 32 GB of memory and on
Nvidia's DGX Spark, and the weights plus a research version with variable loops can be downloaded from
Hugging Face.
- It is also available to try in the Tinker Playground.