◈Bluesky50
YHacker News50
XX / Twitter50
listening to
Jev
150
mentions
tracked
tracked
net sentiment
+25%
48 positive
92 neutral
10 negative
volume & sentiment over time
Sep 25, 2026 · Daily
sources
most discussed
decision / models
active days
1
what people keep raising
- decision / models 36
- built 15
- models 9
- model 7
- open-source / alternative 4
- performance 3
mentions
Showing 26–50 of 50 mentions
-
X X / Twitter 1w ▲ 0 neutralまだあんまりキャッチ出来てないけど、Jevとかをゲーム内のCPUとしてのAIとして使い始めたら何か色々よしなに良さそう まだあんまりキャッチ出来てないけど、Jevとかをゲーム内のCPUとしてのAIとして使い始めたら何か色々よしなに良さそうview source ↗
-
X X / Twitter 1w ▲ 1 positive@BLUECOW009 Jev did better in my tests @BLUECOW009 Jev did better in my testsview source ↗
-
X X / Twitter 1w ▲ 0 neutralDerrière le harness va générer le reste, de la priorisation des tâches et tout. Je penses que Jev peut aider de fou dessus par exemple avec un minimum de coup. L'argent s'aura quoi faire, orchestrer le tout, et manipuler juste tout ça comme une todo app Derrière le harness va générer le reste, de la priorisation des tâches et tout. Je penses que Jev peut aider de fou dessus par exemple avec un minimum de coup. L'argent s'aura quoi faire, orchestrer le tout, et manipuler juste tout ça comme une todo appview source ↗
-
X X / Twitter 1w ▲ 0 neutral@CompleteSkeptic just released this lil interview tool powered by Jev https://t.co/uWv6ci3AKI all the old BYOK apps i wanted to build w/ LLMs are now much more tractable @CompleteSkeptic just released this lil interview tool powered by Jev https://t.co/uWv6ci3AKI all the old BYOK apps i wanted to build w/ LLMs are now much more tractableview source ↗
-
X X / Twitter 1w ▲ 0 positive@noisyb0y1 Jev is genuinely useful @noisyb0y1 Jev is genuinely usefulview source ↗
-
X X / Twitter 1w ▲ 0 positive models📢 TYPESAFE AI · JEV IS NOW LIVE ON https://t.co/kgZeByAQj2 API 🚀 Responds in about 70–500ms, and costs $0.042 per million input tokens (output free), making it a fit for ticket routing, moderation, risk scoring, and agent branching. 🧮 Worth actually reading "70–500ms" once more directly against the newly confirmed GL… 📢 TYPESAFE AI · JEV IS NOW LIVE ON https://t.co/kgZeByAQj2 API 🚀 Responds in about 70–500ms, and costs $0.042 per million input tokens (output free), making it a fit for ticket routing, moderation, risk scoring, and agent branching. 🧮 Worth actually reading "70–500ms" once more directly against the newly confirmed GLM-5.3-FlashX's own genuinely comparable, considerably fast "200 tokens/sec" generation speed, worth appreciating these two figures describe genuinely different things, latency versus throughput, worth being genuinely curious whether a developer building a genuinely latency-sensitive pipeline should prioritize Jev specifically for its own explicit millisecond-level response guarantee over a comparably fast, but not explicitly latency-benchmarked, generative model. 💭 Worth actually reading "output free" once more as a genuinely specific pricing structure worth checking directly against the newly confirmed four-tier discount system, worth appreciating Jev's own genuinely unique, input-only pricing model sits entirely outside that broader percentage-discount framework applied to this account's own more conventional, text-generating models. 🎯 Worth actually reading "Choice, Score, and Noul questions in parallel" directly against the fuller documentation. 👉 Try now: https://t.co/xSnqFyNE3M 🔗 https://t.co/XdOzfnbUkW @justinsuntron @BAI_AGI #TRONEcostarview source ↗
-
X X / Twitter 1w ▲ 0 neutralMario's memory has a UI. My demo links observed jumps, landings, enemy motion and outcomes in a knowledge graph. Observations feed later decisions; you can inspect the evidence. See it in action: https://t.co/HJSalb66rN #JEV #LocalAI AI-assisted. Mario's memory has a UI. My demo links observed jumps, landings, enemy motion and outcomes in a knowledge graph. Observations feed later decisions; you can inspect the evidence. See it in action: https://t.co/HJSalb66rN #JEV #LocalAI AI-assisted.view source ↗
-
X X / Twitter 1w ▲ 0 neutral modelsRecent attentions were on the latest frontier models. ScienceBuddy, JEV, Laya and Kev will shine more soon. Recent attentions were on the latest frontier models. ScienceBuddy, JEV, Laya and Kev will shine more soon.view source ↗
-
X X / Twitter 1w ▲ 0 neutralコメント欄みたけど、たしかに jev って マルチモーダル対応じゃないし、、、 コメント欄みたけど、たしかに jev って マルチモーダル対応じゃないし、、、view source ↗
-
X X / Twitter 1w ▲ 0 negative@typesafeai This is so annoying , today was supposed to be my big JEV implementation & testing day. 🥲 @typesafeai This is so annoying , today was supposed to be my big JEV implementation & testing day. 🥲view source ↗
-
X X / Twitter 1w ▲ 0 positive@spaceman_gilda Welcome! This week's goal for everybody in AI -- show how your stuff works with Jev 🤣 @spaceman_gilda Welcome! This week's goal for everybody in AI -- show how your stuff works with Jev 🤣view source ↗
-
X X / Twitter 1w ▲ 0 positive performanceLooks like Jev from @typesafeai is still the king for now, compared to Kev 4b (now available via @OpenRouter ) since it is a bit too big for my laptop GPU. Kev 4b is certainly getting really close, actually showing better performance on 2 tests. Interestingly Jev via OpenRouter is 4x faster than Jev direct from @types… Looks like Jev from @typesafeai is still the king for now, compared to Kev 4b (now available via @OpenRouter ) since it is a bit too big for my laptop GPU. Kev 4b is certainly getting really close, actually showing better performance on 2 tests. Interestingly Jev via OpenRouter is 4x faster than Jev direct from @typesafeaiview source ↗
-
X X / Twitter 1w ▲ 0 neutral builtC++ devs: your AI output can be an enum. I built JevT++ with typed choices, explicit abstention and an optional Boost.Asio co_await adapter. Local Laya via ONNX/ggml; CUDA optional. MIT: https://t.co/CvpN54FgN5 #JEV #cpp AI-assisted. C++ devs: your AI output can be an enum. I built JevT++ with typed choices, explicit abstention and an optional Boost.Asio co_await adapter. Local Laya via ONNX/ggml; CUDA optional. MIT: https://t.co/CvpN54FgN5 #JEV #cpp AI-assisted.view source ↗
-
X X / Twitter 1w ▲ 0 neutral decision / modelsAn AI can pick the right move for the wrong moment. My JevT++ demo checks model plans against the current game state before execution. Watch the decision graph alongside Mario: https://t.co/HJSalb5yCf #JEV #LocalAI AI-assisted. An AI can pick the right move for the wrong moment. My JevT++ demo checks model plans against the current game state before execution. Watch the decision graph alongside Mario: https://t.co/HJSalb5yCf #JEV #LocalAI AI-assisted.view source ↗
-
X X / Twitter 1w ▲ 0 neutral decision / models@JoshARosen The state-into-decisions framing is why Jev slots under an existing agent graph instead of demanding a rewrite -- most frameworks bolt a decision model on top, this one routes on what the graph already tracks. https://t.co/fD8thmC2ik @JoshARosen The state-into-decisions framing is why Jev slots under an existing agent graph instead of demanding a rewrite -- most frameworks bolt a decision model on top, this one routes on what the graph already tracks. https://t.co/fD8thmC2ikview source ↗
-
X X / Twitter 1w ▲ 0 positive decision / modelsHas anyone trained a Jev / Laya type model on finance data yet? Seems like it would be a no brainer to train it on specific setups for stock / crypto trading. It could act as a layer in an LLM powered quant for quick decision making based on patterns. Has anyone trained a Jev / Laya type model on finance data yet? Seems like it would be a no brainer to train it on specific setups for stock / crypto trading. It could act as a layer in an LLM powered quant for quick decision making based on patterns.view source ↗
-
X X / Twitter 1w ▲ 0 neutraldear vibe coders jev is not generative ai dear vibe coders jev is not generative aiview source ↗
-
X X / Twitter 1w ▲ 0 positive built@shimadaeisuke Loved this, Keisuke. I've featured your post on https://t.co/6rvbCXTB3W, a directory of things people have built with Jev: https://t.co/nJGM0Xe8kk @shimadaeisuke Loved this, Keisuke. I've featured your post on https://t.co/6rvbCXTB3W, a directory of things people have built with Jev: https://t.co/nJGM0Xe8kkview source ↗
-
X X / Twitter 1w ▲ 0 positive built@itIsGokulNair Loved this, Gokul. I've featured your post on https://t.co/6rvbCXTB3W, a directory of things people have built with Jev: https://t.co/nJGM0Xe8kk @itIsGokulNair Loved this, Gokul. I've featured your post on https://t.co/6rvbCXTB3W, a directory of things people have built with Jev: https://t.co/nJGM0Xe8kkview source ↗
-
X X / Twitter 1w ▲ 1 neutral@simplydt @nicochristie I can tell u now jev cannot handle macro lol @simplydt @nicochristie I can tell u now jev cannot handle macro lolview source ↗
-
X X / Twitter 1w ▲ 0 positive modelsWe keep trying to force massive language models into roles that just need fast, deterministic JSON endpoints. A non-autoregressive routing model for typed decisions makes way more sense than burning tokens on things that should just be function calls. #AI #Jev https://t.co/nx4cmwVHwX We keep trying to force massive language models into roles that just need fast, deterministic JSON endpoints. A non-autoregressive routing model for typed decisions makes way more sense than burning tokens on things that should just be function calls. #AI #Jev https://t.co/nx4cmwVHwXview source ↗
-
X X / Twitter 1w ▲ 0 positive built@whosfranki @shadcn Loved this, Franki. I've featured your post on https://t.co/6rvbCXTB3W, a directory of things people have built with Jev: https://t.co/nJGM0Xe8kk @whosfranki @shadcn Loved this, Franki. I've featured your post on https://t.co/6rvbCXTB3W, a directory of things people have built with Jev: https://t.co/nJGM0Xe8kkview source ↗
-
X X / Twitter 1w ▲ 0 positiveI already see Jev, Kev and Lev on twitter. Anyone working on Zev already? 😅 I already see Jev, Kev and Lev on twitter. Anyone working on Zev already? 😅view source ↗
-
X X / Twitter 1w ▲ 2 neutral@painn_x I think we should try the opensource Jev instead 🤣 @painn_x I think we should try the opensource Jev instead 🤣view source ↗
-
X X / Twitter 1w ▲ 0 positiveI run this exact split in production on my own agent. On every message, Jev picks which notes and skills get loaded before Claude sees anything. Measured on 169 real turns: → right note ranked first: 34% → 64% → right note in the top 5: 55% → 78% → 150 to 430ms, $0.00027 a turn Two changes moved those numbers. Each n… I run this exact split in production on my own agent. On every message, Jev picks which notes and skills get loaded before Claude sees anything. Measured on 169 real turns: → right note ranked first: 34% → 64% → right note in the top 5: 55% → 78% → 150 to 430ms, $0.00027 a turn Two changes moved those numbers. Each note is described by the words that should pull it in, not by its title. And Jev sees the previous exchange, so a follow-up like "go test it" still lands on the right file. https://t.co/o2a1K5pAoQview source ↗