DaisyRose-27B
- chriddyp - 2523 sekunder sedanVery excited to see more post-training happening on these open models. Looking forward to trying this one out once it makes its way onto OpenRouter.
Qwen 27B really punches above its weight. 27B is small enough to fit on a Mac Mini / Macbook Pro which is very exciting.
Starting to see more promising examples of companies post-training. From my evals [1], I've tested 3 so far:
1. Fireworks' Ember-1 (postrained on Kimi K3, improved score by 10% in my evals)
2. Particle AI's Particle-1 (fine-tuned on DeepSeek V4.1 Flash, also a 10% improvement and a bit cheaper and faster)
3. Nex AGI's Nex-N2.6 (post-trained on GLM 5.3 Flash. Didn't show an improvement in my eval, and had a couple runaway loops that made it expensive).
With Thinking Machines' western open model release (Inkling Small) a few weeks ago, I expect we'll see more of this, especially out of western companies.
[1] https://plotly.com/blog/nex-n2-6-plotly-data-analytics-bench...
- woah - 32486 sekunder sedanThe heavy reliance on eval datasets showcased here and also evidenced by the frontier lab's high payments for such data cast doubt on the claims that AI will become self-improving soon
- pkoiralap - 27234 sekunder sedanSabi-noema also looks like a close competitor in the same space. Built from qwen3.8 and made better.
https://huggingface.co/sabicap/sabi-noema
Also HF model here: https://huggingface.co/RadixArk/DaisyRose-27B
- jacquesm - 31533 sekunder sedanI always wonder what you lose when you start maxing out some specific benchmark. It is really hard to do this in a balanced way and I would love if everybody that did these kind of efforts would take the time to also list the before/afters of a number of other benchmarks that they did not use in their training.
- xlayn - 34755 sekunder sedanSnap, no gguf to download, will wait for one to appear to test, any improvement over an already good model is good news.
"18 days on 144 H200 GPUs" insane.. didn't thought so much hardware would be needed to train Qwen.
- mips_avatar - 25648 sekunder sedanHow did you guys set up the RL topology? Were you guys doing individual RL nodes (4xh200 rollout/4xh200 train) or did you have dedicated trainer and rollout nodes?
- octoberfranklin - 34239 sekunder sedanFWIW that is about $331,000 on low-onboarding-grief clouds (e.g. vast.ai) right now using the SXM variants they used.
You might be able to shave that down using a different flavor of H200 (like PCIe) but it would take more hours, so benefits are unclear.
- 0x69420 - 17606 sekunder sedan3.8 27B is an interesting choice. on the face of it, doing RL on it feels like refrying twice-cooked pork then stuffing it in the microwave for good measure. the fact that even one benchmark (tb4.1) increased more than marginally is kind of miraculous. would be interesting to see where "miles" would take eg. 3.8 flash next, which is already impressive despite being an under-trained architectural preview.
- robertkarl - 35145 sekunder sedanI look forward to testing this against my private eval set to determine if it's benchmaxxing or actual improvement based on just 1200 tasks. downloading it now
- XCSme - 34723 sekunder sedanIs it post-training or bench-maxxing?
- - 37574 sekunder sedan
- amarinhodev - 29398 sekunder sedan[flagged]
- bzvox - 3224 sekunder sedan[dead]
Nördnytt! 🤓