Observed Signal · Oct 5, 2026 · Technical Release · Source: The Product Compass · Impact: 2/5 · Sentiment: Positive
TypeSafe's Jev Decision Model Explained for Product Managers
The article introduces Jev, a decision model from TypeSafe AI, released in September 2026. Unlike LLMs, Jev returns a decision with a probability based on input text and a question, supporting operations like Noul (yes/no), Choice (one of many options), and Score (rating). It processes up to 32K tokens, with free output tokens and input at $0.042 per million tokens. The author tested it against other models, finding it cost-effective and accurate. Jev is integrated into AskOne for moderation, with rules encoded in prompts to handle abuse and prompt injection. The article notes Cloudflare released similar models (Clef) on October 1, and provides guidance for PMs on quick wins, evaluation, and setup templates.
Introduces a new AI model type that could lower the cost and complexity of implementing AI-driven decision features in products, with direct implications for AI in martech.
Track TypeSafe AI Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- TypeSafe AI released Jev in September 2026.
- Jev processes text and returns decisions with probabilities.
- Input cost is $0.042 per million tokens; output tokens are free.
- Jev supports up to 32K tokens per request.
- Cloudflare released Clef decision models on October 1, 2026.
Connected Companies & Entities
2 Entities mapped“Jev is a decision model from TypeSafe AI, released in September....”
“On October 1, Cloudflare released its own decision models, Clef and Clef-flash....”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Jev Decision Model: Fast, Cheap Classification for AI Pipelines
Claire Vo, founder of ChatPRD, demonstrates the new Jev decision model from TypeSafe AI in a video walkthrough. Unlike standard LLMs that generate text, Jev returns type-safe structured values (choice, score, probability) at a cost of $0.04 per million input tokens with no output charge. Vo details five real-world applications: categorizing 1,700 PRs for $0.09, analyzing local Claude Code and Codex sessions, triaging Gmail, building a product insights graph from 1,100 signals with 200,000 classifications, and creating a live audience dashboard from 4,500 YouTube comments. She emphasizes combining Jev's fast, cheap classification with more capable models like GPT-6 Astra for analysis and generation, achieving cost-effective and performant AI workflows. The video also demonstrates a real-time voice-to-color emotion mapping app, highlighting Jev's low latency for interactive use cases.
TypeSafe Launches Jev, a Fast Decision Model for AI Systems
TypeSafe AI, founded by former OpenAI researcher Diogo Almeida, launched Jev, a 'System One' model for fast, structured decision-making. Jev returns typed probabilistic outputs with confidence scores for predefined choices, avoiding text generation and thus eliminating token-by-token decoding, hallucinations, and the need for validators. Trained with a novel RLCD technique, it offers well-calibrated, confident decisions. Jev is 20-200x faster and 40-400x cheaper than frontier LLMs, with response times of 70-500ms, free output tokens, and input costs of $0.042/M. The launch gained massive attention, but integrations with Vercel, Cloudflare, LangChain, and others have solidified developer interest. TypeSafe raised $40M seed funding led by DCVC, valuing it at ~$200M. Now available without a waitlist, Jev is used for ad analysis, agent reasoning, and on-chain trading, with Vercel AI Gateway offering free access until September 25.
Exploring Jev for AI-Driven QA with agent-device
New article: Jev is a model from TypeSafe that chooses between predefined options instead of generating text. We paired it with agent-device to explore what that approach could mean for AI-driven QA: reading app state, selecting actions, and checking the result. We walk through our proof of concept, how the two tools work together, and a recorded run that took 14 seconds and cost $0.0023 in model inference.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.