{"slug": "deep-llm-jev-users-report-10x-faster-and-54-5x-cheaper-than-the-models-they-only", "title": "Deep|LLM: Jev Users Report 10× Faster and 54.5× Cheaper Than the Models They Replaced; Only 3.7% in Production", "summary": "An analysis of 6,277 public discussions from the first week of TypeSafe AI's Jev, a non-generative \"decision model\" released September 15, 2026, found that users report a median 10× speed-up and 54.5× cost reduction versus the models it replaced, with median latency of 300 ms versus 2,924 ms in 16 head-to-head cases. Accuracy gains were marginal — Jev won 10 of 17 head-to-head comparisons with a median 1.6 percentage-point gap — and only 80 of 2,153 hands-on cases (3.7%) were in production, with indie developers (35.9% of identified authors) dominating discussion over large-company engineers (2.7%).", "body_md": "Jev is a “decision model” from TypeSafe AI, released September 15, 2026 and opened to all users on September 20. It does not generate text. It answers questions with a fixed answer set: pick an option, score on a scale, or judge true/false, and attaches a confidence score. The launch quickly gathered industry interests, and some investors were asking whether it’s a significant negative to compute demand. **As we addressed in our report earlier, we disagree with that concern and believes Jev is more of an interesting trial with limited impact on LLM.**\n\n**To analyze Jev further, we decided to have a deep dive into what Jev use cases are really about. This note covers 6,277 public discussions and use cases from the first 7 days; 2,153 are from people who actually used or tested it.**\n\n- **Demand sits on fast decisions with a fixed answer set.** No single use clears 20%. Of the 1,284 cases with an identifiable use, the largest groups are real-time control in games, robots and simulations (18.8%), agent control decisions (16.1%) and content classification (15.7%).\n- **Indie developers dominate the conversation; big-company engineers barely show up.** Of the 2,140 authors whose role we could identify, 35.9% are indie developers, 23.4% are AI creators and KOLs, and just 2.7% are engineers at large companies.\n- **Speed: 10× faster than the model it replaced or was tested against.** Median user-reported speed-up is 10× (n=72): 10× vs frontier models, 5× vs small models. In the 16 cases with latency for both Jev and the prior system, Jev’s median is 300 ms vs 2,924 ms. The vendor’s 193.6× is a peak against the most expensive model.\n- **Cost: 54.5× cheaper than the comparison model; the saving depends on what it replaced.** Median user-reported cost multiple is 54.5× (n=56): 188× vs frontier models, 17× vs small models. The vendor’s own comparison with GPT-5.6 Terra is about 76×; the 444.6× in marketing is a peak against the most expensive model.\n- **Accuracy: Jev and the systems it replaced each win some head-to-heads; gaps are small.** In the 17 cases with accuracy for both, Jev is ahead in 10 and behind in 7; median gap is 1.6 percentage points. Of 241 cases that assessed accuracy, 83 rated Jev better and 65 worse.\n- **Jev’s confidence scores miss by about 10 percentage points on average, and run clearly high on unfamiliar rating questions.** Median user-measured ECE (expected calibration error: average gap between stated confidence and actual accuracy; 0 is perfect) is 0.097 (n=27). An independent test on unfamiliar tasks found 0.107 overall, but 0.325 on rating questions, where Jev was right only 44.7% of the time.\n- **Developers put cheap small models next to Jev almost as often as the strongest large ones.** Of the 432 cases that name a comparison model, 48.6% mention open or small models and 59.7% mention frontier models.\n- **Criticism is common.** Abandonment after trying Jev is not. 25.1% of all 6,277 cases contain criticism and 40.6% contain praise, but only 1 of the 2,153 hands-on cases ended with Jev being dropped.\n- **Production use is still rare.** Most activity is experimental. 80 of 2,153 hands-on cases (3.7%) are in production; prototypes, side projects and trial demos make up 62.4%. Seven days of data: treat this as a baseline, not a run-rate.\n\n## 1. What developers use it for\n\nIn week one, public discussion was mostly opinion; few people had actually used or tested Jev. Opinion and analysis are 41.8% of the 6,277 counted cases (Figure 1). The 2,153 hands-on cases are the better read on demand; the rest of this note uses that cut.\n\n*Source: FUNDA*\n\nAmong hands-on cases, the largest uses are real-time control, agent control decisions and content classification. None reaches 20%. They are 18.8%, 16.1% and 15.7% of the 1,284 hands-on cases with an identifiable use (Figure 2). Table 1 shows a typical case for each.\n\n*Source: FUNDA*\n\nAn open-source browser automation project uses Jev at every step to decide what to do next and which control to click or fill, and calls a small model only when it needs to type. One flight search took 7.1 seconds end to end: that is whole-task time, not a single decision. “Give it one goal. TypeSafe’s Jev picks an operation and an element. A small LLM writes text only when the operation is TYPE_TEXT”\n\nIn real-time control, someone built a Tesla self-driving control demo with Jev in under an hour. Build time only; no other figures. Still a demo.\n\nShared traits: preset answer list, frequent calls, speed matters. In the 2,153 hands-on cases, Choice (pick one of the given options) is the most common question type at 23.0%; Noul (true/false) and Score (scale) are less common (Figure 3). Against the vendor’s own use list, real-time applications and branching decisions in code are 28.9% and 27.8% of the 1,284 identifiable-use cases. Bulk data processing is 13.2%: not a main use today.\n\n*Source: FUNDA*\n\n## 2. What Jev replaces\n\nJev does the deciding inside a program, not the writing. Classification and routing is the largest group among the 1,284 hands-on cases with a clear purpose (27.0%), then real-time and games, then search and ranking (Figure 4). These steps choose from preset answers and run many times; text generation does not matter there.\n\n*Source: FUNDA*\n\nMost of what Jev replaces is frontier models, though more than six in ten hands-on cases never say what came before. Among the 2,153 hands-on cases, prior approach is unknown in 62.1% and a frontier model in 15.1%. Rules or heuristics (keyword matching and similar) are 6.6%; human review is 0.7% (Figure 5). When Jev’s confidence is low, the most common fallback is a person (7.4%); 7.8% have no fallback at all (Figure 6).\n\n*Source: FUNDA*\n\n*Source: FUNDA*", "url": "https://wpnews.pro/news/deep-llm-jev-users-report-10x-faster-and-54-5x-cheaper-than-the-models-they-only", "canonical_source": "https://fundaai.substack.com/p/deepllm-jev-users-report-10-faster", "published_at": "2026-09-24 13:52:28+00:00", "updated_at": "2026-09-24 14:02:19.332321+00:00", "lang": "en", "topics": ["large-language-models", "ai-products", "ai-tools", "ai-startups"], "entities": ["TypeSafe AI", "Jev", "FUNDA", "GPT-5.6 Terra"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/deep-llm-jev-users-report-10x-faster-and-54-5x-cheaper-than-the-models-they-only", "markdown": "https://wpnews.pro/news/deep-llm-jev-users-report-10x-faster-and-54-5x-cheaper-than-the-models-they-only.md", "text": "https://wpnews.pro/news/deep-llm-jev-users-report-10x-faster-and-54-5x-cheaper-than-the-models-they-only.txt", "jsonld": "https://wpnews.pro/news/deep-llm-jev-users-report-10x-faster-and-54-5x-cheaper-than-the-models-they-only.jsonld"}}