Clef: Open-weight decision models, and new RL fine-tuning platform
jasondavies
474 points
170 comments
October 01, 2026
Related Discussions
Found 5 related stories in 103.3ms across 8,245 title embeddings via pgvector HNSW
- I built non-autoregressive decision models with RL a year ago nandakishor_ml · 1169 pts · September 19, 2026 · 59% similar
- Does an open-weight decision model beat a hosted one? Jev vs. Laya Jmeg8r · 11 pts · September 22, 2026 · 54% similar
- Inkling: Our Open-Weights Model vimarsh6739 · 850 pts · July 15, 2026 · 50% similar
- A $500 RL fine-tune of a 9B open model beat frontier models on catalog review ilreb · 98 pts · July 28, 2026 · 50% similar
- Show HN: I RL-trained an agent that trains models with RL (for ~$1.3k) Danau5tin · 101 pts · July 14, 2026 · 50% similar
Discussion Highlights (20 comments)
warkdarrior
Can someone explain how so many folks managed to build decision models within days or weeks after Typesafe came out with Jev? Is this concept of decision models been in the works for a while? Is it easy to copy?
johnecheck
Wow, Cloudflare is definitely buying some goodwill from me. Just consistently interesting new releases alongside and solid products at great prices. Seems nearly too good to be true.
hbcdbff
“Urgency” of “yes”?
aryabakh
it's great to see Cloudflare releasing consumer edge level models.
ssiddharth
Pricing is $0.24/million input tokens which is ~6x compared to Jev. Clef-flash is at $0.09 which is way more competitive.
DesaiAshu
brb while I build my entire cloud stack on Cloudflare
open592
2 years in stealth...
swingboy
It allows image input. Nice!
yipinwong
A question someone not trainined in AI/ML field, Is a decision model that easy to crete that there are floods of these JEV alternatives already? Or are companies/people already building this based on say an arXiv docs? n --- The pricing is ... hm more expensive but not at the point I won't give it a try due to the embeded vision encoding
zwaps
No mention of calibration. Is it just another llm finetune?
bityard
Clef is based on Qwen3.8-27B and Clef-flash is based on Qwen3.8-9B (edit: actually Qwen3.5-9B). So, similar in spirit to Kev by my understanding, but based on a newer model.
MisterMunchkin
Imagine making your whole company on one model and then being cucked by everyone within a week. I don't think I've ever seen anything like it.
6thbit
I wonder if a good usecase for this would be cloudflare's WAF rules. Give broader request context to the decider and let it pick type of challenge/block traffic directly. Perhaps that may be too costly atm
manlymuppet
Am I hearing this right, that they made a decision model based on Typesafe's new paradigm, and actually made a model better than Jev based on Typesafe's own ranking? And it's only been a few weeks.
buildbuildbuild
Open weights, not open source. The weights have permissive licensing, but the data and training pipeline are not published to reproduce them from their proprietary Qwen starting points. Weights are not "source."
ksymph
With all these new Jev-like models popping up, has anyone actually started building anything with them yet? It's odd how quickly they've multiplied despite being relatively niche in their use cases, as far as I can tell. I suppose they're simple and cheap enough to make that it's a sort of 'why not' thing for a lot of these companies.
schainks
AMAZING, thanks, Cloudflare!
esafak
I feel bad for the Jev guys. I wonder if they anticipated this much competition?
selfawareMammal
Horrible name
mococa
I bet this's 100% slop.