Introducing Bespoke Nimble: an open data, open model, open recipe for an open Jev. Code and info: Model: Data: * A new data curation recipe called contrastive data curation. * Slightly change facts to generate negative data. This pushes the model to discriminate better and become a better decision maker. The calibration is implicit. * Didn't do ablations but I think this is a critical piece! * This also means training data doesn't need probabilities. * Data covered 10 categories, and is fully synthetic. * This data is split into train and eval. Training * LoRA finetune of Qwen3.5-9B. * Distillation-free: we use Jev to only evaluate. * No RL yet! Serving * Parallel constrained decoding as suggested by @NielsRogge and @harshagundal. Results: * The post-trained Qwen (Nimble) became substantially better on our curated eval: 66% for Qwen to 90% for Nimble. Jev is at 93%. * 100ms on H100 and free to use on your macbook! Feel the AGI for free. * 2 days of building in public. :) Big caveat is that there is no standard benchmark to measure performance, and it's possible Nimble is much worse on other benchmarks compared to Jev. But it should be better than Qwen! We thank @typesafeai for making Jev and the inspiring discussions in the community. Hope this release lifts all the boats and encourages more research and activity in this space.

I am reading the exact opposite way. This effort is very cool, and isn't it kind of amazing to collect your own data, run a custom midtrain, and then running your own RL to end up at the same performance as Astra max for 1/2 the inference price? Imagine what can then be next.

Hi mathematicians, don't fret. AI beat humans long long ago in chess, yet we are mostly interested in watching humans play chess. Likely something like that will play out for you as well. We should still have olympiads. We should continue to have conferences where mathematicians go talk. I will still be interested in hearing what Terrence Tao has to say more so than an AI smarter than him that has something to say.