The rise of decision models
Vendors are racing to ship small, dedicated classifier models alongside their general LLMs, and competitors are matching each other's releases within days, not quarters.
-
Small, dedicated decision models are emerging as their own category alongside general LLMs, and vendors are matching each other within days
Sebastian Raschka wrote Sept 29 that TypeSafe AI's classifier model 'Jev' became 'a cultural phenomenon' in two weeks, beating general LLMs on cost and speed for classification work. OpenAI's Nikunj Handa said its own Decisions API was built in about a week after Jev 'nerd sniped' OpenAI's engineers, per swyx's Latent Space. Unsloth and llama.cpp both shipped runtime support for rival decision models, Laya and Clef, within days of each other. Raschka · Latent Space · Unsloth releases
-
Amazon released an open-source 2B decision model, Strands Decider, joining the vendor race in small classifiers - single source. Tech Insider
-
SHIPCloudflare and Amazon both ship open decision models the same day
Cloudflare released Clef and Clef-flash, Apache-licensed classifier models built on Qwen, the same day Amazon's Strands Labs open-sourced its own 2B decision model with training data included - rival vendors are now shipping these models within days of each other, not quarters. DataNorth · Digital Applied
-
WATCHCloudflare's new Clef decision models get day-one GGUF support
Independent quantizer bartowski shipped GGUF versions of Cloudflare's Clef and Clef-flash classifier models within a day of release, a sign the fast-cloning decision-model category is getting real community tooling, not just vendor benchmarks. bartowski
-
SHIPMicrosoft ships Decision-1, undercutting the week-old decision-model market
Decision-1 is a post-trained version of Alibaba's open-weight Qwen3.5-9B, returning a probability across fixed options instead of open text. It's live in Microsoft Foundry at $0.042 per million input tokens with free output, and Microsoft claims it beats GPT-6 Sol on latency by 35x. TestingCatalog · RuntimeWire
-
Unsloth's new decision-model trainer reportedly lifts a small Qwen model's accuracy from 21% to 74%, vendor-reported. AlphaSignal
-
SHIPTypeSafe AI raises $870 million at a $7.5 billion valuation
Andreessen Horowitz led the round, with Sequoia and DCVC joining, three weeks after TypeSafe's Jev decision model launched - the clearest sign yet of investor appetite for the new decision-model category. Pulse2
-
Hamel Husain and Antaripa Saha published a practical guide to tuning decision-model thresholds for evals, Oct 20. Hamel Husain, How To Use Decision Models for Evals
-
Decision models are consolidating into a real product category, not one vendor's novelty - three major labs have converged on nearly the same design within six weeks of each other.
Google shipped its Gemini Decisions API and Gemini Decision 1 model Nov 11, closely modeled on TypeSafe's Jev - it even reuses Jev's term 'Noul' for yes/no questions - at OpenAI's same 10-cents-per-million-token price, per Simon Willison's same-day hands-on test. He notes Google now joins Jev and OpenAI's Decisions API as a genuine third entrant in a category barely six weeks old. Simon Willison, Gemini Decisions API
-
Decision models, cheap APIs that answer yes/no and rubric questions, are becoming a standard model category with several vendors.
Simon Willison's November 11 post covers Google's Gemini Decision 1, which he calls almost identical to Jev's API. It takes text, images, audio, video and PDFs, costs 10 cents per million input tokens with free output, and returns probabilities where Jev and OpenAI returned 1.0 or 0.0. Liquid AI released its d1.2 decision models on November 16, per The Neuron. The category forming across vendors is my reading, not a claim in either source. Willison, Gemini Decisions API · The Neuron, November 17 digest