← all stories
Jev
Jev is a judgment model from TypeSafe AI that scores difficulty and accuracy, released on September 15, 2026. Its use has expanded from TypeSafe AI's router on OpenRouter, available since September 28, 2026, to a new local execution app called Ollaya.
Synthesized from 2
Yomimono stories · updated
2h ago
Jev is a judgment model from TypeSafe AI. It outputs judgments and confidence scores for multiple options. TypeSafe AI released it on September 15, 2026. The company is led by CEO Diogo Almeida, who contributed to the development of ChatGPT.
On September 28, 2026, TypeSafe AI's Jev Router became available on OpenRouter, the AI gateway service. The router sends a prompt through Jev to score the task's difficulty and accuracy, then assigns an appropriate AI model. Easy tasks go to inexpensive small models; difficult tasks go to more expensive state-of-the-art models. OpenRouter's earlier routers analyze task type but not difficulty. Testing across four AI agent benchmarks found Jev Router completed 82 percent more tasks than OpenRouter's Auto Router. Jev Router does not save prompts and does not use them for training.
On October 4, 2026, Ollaya appeared as a new execution app that runs Jev-style decision models on local hardware. It supports Windows, macOS, and Linux and is published as open source under Apache License Version 2.0. A listed example has the decision model Winnow-E4B read a message about a third double charge and return Intent: refund request (91%), Urgency: present (92%), and Anger level: 2.89 (86%). A chart comparing Jev's API service against a local run on a GeForce RTX 4070 shows that, depending on the model, local accuracy gets fairly close to Jev's while latency is shorter. Ollama added decision-model support in version 0.35, covering three models (nimble, tev1, and tev1:0.8b) and allowing a maximum of 15 options per request. Ollaya can run higher-performance models and set up to 255 options.
Key facts
- Model type
- Jev is an AI model that outputs judgments and confidence scores for multiple options. ↗
- Release date
- TypeSafe AI released Jev on September 15, 2026. ↗
- Company
- TypeSafe AI is led by CEO Diogo Almeida, who contributed to the development of ChatGPT. ↗
- Router availability
- TypeSafe AI's Jev Router became available on OpenRouter, the AI gateway service, on September 28, 2026. ↗
- Router benchmark result
- Testing across four AI agent benchmarks found Jev Router completed 82% more tasks than OpenRouter's Auto Router. ↗
- Router privacy
- Jev Router does not save prompts and does not use them for training. ↗
- Ollaya platforms
- Ollaya supports Windows, macOS, and Linux. ↗
- Ollaya license
- Ollaya is published as open source under Apache License Version 2.0. ↗
- Ollaya example output
- The decision model Winnow-E4B reads a user message about a third double charge and returns Intent: refund request (91%), Urgency: present (92%), and Anger level: 2.89 (86%). ↗
- Option ceiling
- Ollaya can set up to 255 options, while Ollama version 0.35 allows a maximum of 15 options per request. ↗
Timeline
Synthesized by Yomimono from the cited Yomimono stories below, each itself
sourced, then editorially reviewed. Every
fact links the story it came from.
Facts
- Release
- game · 2026-10-14 · 2026-10-15
- Release
- game · 2026-09-15 · 2026-10-03
Connections
Structured graph also available as JSON at /public/entities/jev.
CC BY 4.0.
All coverage
2h ago
TypeSafe AI released Jev 1.14.0 on October 14, 2026, adding image input and a way to raise context length to 262k tokens per request. The decision model, which picks from preset options with a confidence score rather than generating text, had multimodal support disabled because non-text input performed worse. Long context stays off by default. The team also posted self-run comparisons showing higher intelligence, better cost efficiency, and a lower error rate than its previous model and other decision models.
5d ago
Opper published jevman, an open-source benchmark that has AI models play Pac-Man to compare decision-making. Six models, including Jev, Kev, Laya, Clef, and GPT-6 Luna Decisions, each play 100 games against standard-rule ghosts. Models receive maze, dot, and ghost data as JSON and return up, down, left, or right as probabilities within a 2-second limit per junction. Ranking uses the 100-game average score, with a 95 percent confidence margin, so models inside that margin are treated as tied.
Oct 4
A new execution app called Ollaya runs Jev-style decision models on local hardware. Jev, which appeared on September 15, 2026, is an AI model that outputs judgments and confidence scores for multiple options. Ollaya supports Windows, macOS, and Linux, and is published as open source under Apache License Version 2.0. In a listed example, the decision model Winnow-E4B reads a user message about a third double charge and returns Intent: refund request (91%), Urgency: present (92%), and Anger level: 2.89 (86%). A chart comparing Jev's API service against a local run on a GeForce RTX 4070 shows that, depending on the model, local accuracy gets fairly close to Jev's while latency is shorter. Ollama added decision-model support in version 0.35, but that release covers three models, nimble, tev1, and tev1:0.8b, and allows a maximum of 15 options per request. Ollaya can run higher-performance models and set up to 255 options.
Oct 3
TypeSafe AI's Jev Router became available on OpenRouter, the AI gateway service, on September 28, 2026. The router sends a user's prompt through Jev, the judgment model TypeSafe AI released on September 15, 2026, to score the difficulty and accuracy of the task, then assigns an appropriate AI model. Easy tasks go to inexpensive small models and difficult tasks to more expensive state-of-the-art models, cutting the waste that comes from routing easy and difficult work to the same model. OpenRouter's earlier routers analyze the type of task but not its difficulty. Testing across four AI agent benchmarks found Jev Router completed 82% more tasks than OpenRouter's Auto Router. Jev Router does not save prompts and does not use them for training. TypeSafe AI is led by CEO Diogo Almeida, who contributed to the development of ChatGPT.