Jeff: 0.8B decision models trained at home that match Jev on classification
Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms
Jeff is a family of tiny fine-tuned models (Qwen3.5 0.8B/2B, Gemma 4 E2B) that turn plain-language descriptions and options into calibrated probabilities in a single forward pass — about 22 ms on an RTX PRO 6000 and 28 ms on an M4 Max. Trained entirely on local hardware with synthetic data, they approach or beat Jev on classification and grounding benchmarks, though they trail on reasoning-heavy tasks. A short fine-tune can boost held-out accuracy from 31.7% to 95.8%.
Small models don't reason. Expect fast, calibrated choices between the options you describe, not multi-step reasoning.
- adrithmetiqa
Forgive my lack of understanding but how long before Jev type functionality is just built straight into all frontier models?
- AgentMasterRace
I compared it to Jev in my current use cases and it's very inaccurate. 70% vs 94% . for classification, it's unacceptable.
- trebligdivad
What proportion of commercial LLM use is classification? I'm just wondering what happens to business AI spending/data centre usage when they realise they don't need full LLMs.
- imranq
Everyone saying you could replace Jev or decision type models with an LLM with bolted schema output constraining are missing the point completely. Its about extreme speed and cost effectiveness with high quality, neither of which you are going to get with LLMs even with these KV-cache tricks
- phlipski
For those of a certain age - the fact that Askjev.com is still available astounds me.
- velominati
Typesafe has been quite about the underlying technology behind Jev. Given the speed and cost my hypothesis is that it doesn’t input tokens the way that LLMs do, ie iterating over every word and drawing the connections between each. That is an o(n^2) problem which is why LLMs are so expensive as they scale.
- qtalen
Awesome, I was just looking for a decision-making model that can be deployed locally, and here you are. Thanks a lot!
- danbrooks
This type of project looks extremely useful. There was a lot of buzz around Jev, but having models that run locally and can be fine-tuned is extremely helpful.