Eschewing language does a few things: It makes the model incredibly cheap and fast, and because users define the outputs in advance, it cannot hallucinate. Its output tokens are free, and input tokens are metered by the billion, not the million<\/p>\n Developers are taking a great interest in the product; the company briefly lost the ability to serve users from its API because demand was so high. Jev appears most useful for software automation. Thus far, software developers see it as a cheaper and more robust way to incorporate intelligence into their code<\/p>\n For example, Pranit Sharma, a software engineer at Vercel, a company making agentic infrastructure, said his company had used OpenAI\u2019s ChatGPT Luna 5.6 to run a classifier to review commands for safety. When Vercel replaced OpenAI\u2019s Luna with Jev, it got results five to 18 times more quickly and with greater accuracy<\/p>\n Another developer, Bryo AI CTO Nikhil Mudholkar, tested<\/a> Jev against Gemini for classifying business emails. In his test, Gemini was slightly more accurate, but 10 to 20 times more expensive. More interesting to Mudholkar were Jev\u2019s confidence scores \u2014 \u201cit is the only one that hands back a real probability which makes it ideal for automating workflows!!\u201d<\/p>\n Besides replacing LLMs in certain use cases, the new model can also augment them, acting as a smart check on misbehavior. Using agents to monitor agents can quickly become expensive, but using Jev to do so, Almeida argues, makes sense. He sees users deploying Jev to track LLM agent traces and prevent jailbreaks<\/p>\n \u201cAt the end of the day, it delegates the hallucination problem a little bit to the user,\u201d explained Armin Ronacher, the CTO of Earendil, which builds the open omes back with 50% probability, maybe this is a coin toss, and I disregard it. But if it\u2019s 95%, sure, then I can do something with it.\u201d<\/p>\n Another potential use for Jev is model routing, Ronacher said. Predicting whether a given workload requires<\/a> a specific model would be useful, but using an LLM for the job would be expensive. Jev\u2019s low cost and speed make that kind of real-time sorting possible<\/p>\n And that\u2019s Almeida\u2019s hope. The model is named after William Stanley Jevons, the 19th-century economist whose eponymous paradox describes how the falling cost of a commodity can lead to it being used more and more. In this case, the falling cost of intelligence should lead to its widespread deployment<\/p>\n \u201cWe think that there\u2019s just going to be smart software all over the place in a way that\u2019s emergent and distributed \u2026 much more like the early internet than you know<\/a> like the mega apps that people are trying to build right now,\u201d Almeida said<\/p>\n Almeida is tight-lipped about the model\u2019s architecture, which outside observers suspect is built on top of an open-weight LLM. The company refers to Jev as a \u201cSystem One model,\u201d focused on intuition rather than reasoning, and specifically focused<\/a> on the right task. Almeida says Jev is trained exclusively on synthetic data using a technique he calls \u201creinforcement learning from calibrated decisions.\u201d<\/p>\n \u201cWe made an early bet that we will be making all of our data, and that has been one of the best bets I\u2019ve ever made in my life \u2014 better than our launch, in my opinion, better than RLHF,\u201d he told TechCrunch. \u201cHalf of [our company] is a lab that basically owns this entire subfield of statistically well-understood synthetic data, and that is now my life joy.\u201d<\/p>\n For now, Jev stands alone as this kind of model, but Ronacher expects that competitors will spring up now that its utility is apparent<\/p>\n \u201cWe should have seen this earlier in many ways, but presumably because the LLMs are so cheap and subsidized, you often don\u2019t have to be creative yet,\u201d he said<\/p>\n TypeSafe itself will be building more versions of the model, in new modalities. Asked if TypeSafe is a frontier lab, Almeida said, \u201cthe main product of frontier labs is fear or hype. I would like our main product to be intelligence\u2026[but we are] not a lab in the sense of, you know, like bet on infinite wealth, or a religion, or building God in a data center, or whatever is the thing of today.\u201d<\/p>\n \n Related:<\/strong> \n <a href="https:\/\/yoursite.com\/automation-africa\/" title="WhatsApp Marketing<\/a> Automation Africa: 6 Dangerous Mistakes Brands Make in Nigeria”> Join Justfine Infotech<\/strong> and build real digital skills in AI, automation, web development, digital marketing, office productivity, e-commerce, freelancing and cybersecurity.<\/p>\n Available Programmes:<\/strong> WhatsApp:<\/strong>![\"\"](\"https:\/\/justfineinfotech.com\/wp-content\/uploads\/2026\/09\/Screenshot-2026-09-18-at-12.08.01-PM.png\")
\n
\n Digital Automation Training Benin: 5 Winning Skills Employers Demand in 2026
\n <\/a>\n <\/p>\n
\n WhatsApp Marketing Automation Africa: 6 Dangerous Mistakes Brands Make in Nigeria
\n <\/a>\n <\/p>\n<\/div>\nWant to learn this practically?<\/h3>\n
\n 6 Weeks Certificate \u2022 3 Months Professional Certificate \u2022 6 Months Diploma \u2022 Full Professional Diploma<\/p>\n
\n +229 01 57 57 99 15
\n +229 01 66 68 11 60<\/p>\n