Today we’re announcing our “ThinkingCap” efficient model series with a 2× thinking token reduction on average in Qwen 3.6 27B, with up to 10x faster generation on individual examples.

The results across seven out-of-domain benchmarks: 🏆 Up to 10x faster in specific responses 📉 ~46% fewer reasoning tokens on average 🎯 −0.7pp accuracy difference (comparable accuracy) 🔄 Fewer reasoning loops & dead-ends ⚡ Lower latency, lower inference cost, better real-world throughput

I am using this improved model in my local cluster “Bob” for some time now. I’s fast and also shortening the answers without degrading the output.

Our goal was to get to same answer sooner with fewer tokens burned and without loosing accuracy.

🔗 Full technical write-up with methodology & numbers in our blog post: 🔗 HuggingFace LINK:

PS: Efficiency not just in open models is our core thesis since we started year ago. If you want to work on frontier of efficiency, join us!

Replicated on Czech-language RAG (tender docs, 3090, same quants): −29–55 % thinking, 0 quality loss, and 3× lower variance in thinking length across seeds — more predictable latency in production. Fits 24GB. Well done 👏

🔥

🏎️ love it!

🙌🔥

;)

Interesting.

your 🤖 Henry would love this model!

Any plan for official mlx ?

Qwen 3.6 36b-A3 will work also?

Our training algorithm will work for most open models.

i will wait for mtp, this is currently slower

Good stuff

Great idea, similar to autothink - ?

what makes the 2x token reduction so remarkable is it's a direct result of rethinking compute and dataflow from scratch

Interesting 👍 Does it support MTP?

Yep

Can I make Beat Saber 2 with it? 🙃✌️

Go ahead! ;)

Pitty it's at the beginning of summer holidays, tried to promote it to a community neewsfeed but everyone is gone for weeks :(. But thanks for making europe great again 😁 keep rocking

Great job, @JaroslavBeck ! Keep up the excellent work. You are truly inspiring, and the Czech Republic appreciates what you do more than it may seem at first glance. Thank you to your entire team.

Really cool guys!! 🇨🇿🇨🇿🔥 I have 4 questions below👇