X · Laya
Laya on X
The model itself shipped as a repository and a paper, so the explaining happened in public — and the arguing did too. These are the posts worth a reader's time: the port that carried the model further than anything the publisher wrote, benchmarks run by people with no stake, the sceptics, and the one test that came back with a loss.
28 posts described Likes read 2026-09-27 The projects behind them: builds
- X
介绍比Jev快50倍,在你设备上跑的laya-mlx! 只在你的设备上占用最高1G内存 Laya是一个开源的类似于Jev的,基于文本输出概率的分类系统 我将其移植到MLX,并且做了一些性能优化! 视频中就是这个模型在我的本地M3Max上玩贪吃蛇 这个模型能够以每秒决策60次的速度玩贪吃蛇! https://t.co/31KGUiNunb https://t.co/5kAXnrrdUA
The post that carried this model further than anything the vendor published: 14,000 likes for a port to MLX and a Snake game running at 60 decisions a second on a laptop. Read it for the port and the memory ceiling; the speed comparison in the text is against the hosted model's API-side numbers.
- X
Local Laya moggs Jev at @grok 4.7-built Tetris 🧩 An open-weights System One model called Laya, beat cloud-based Jev at playing Tetris by making decisions 11 times faster, running locally on a 16GB MacBook Air! Run AI models locally -> https://t.co/RbcCOIgVkj…
A Tetris build against a cloud model, local Laya winning by making decisions eleven times faster on a 16 GB MacBook Air. It is a game, so the result is about the loop rather than the model, and the post does not pretend otherwise.
- X
poor guy claim to have built Jev a year ago but no one cared, and now Jev stole all the thunder many people are saying “you gotta tell your story” or “marketing is important”, and they just completely missed what actually made the difference here i just looked into this laya…
A thread that starts as a joke about a year-old project nobody noticed and turns into the sharpest take in the corpus: the difference was not marketing but that Jev framed the problem as a category. Worth reading for the argument about framing rather than for anything about the weights.
- X
JevBench results are in. Jev still in the lead, but it's close. https://t.co/8P4lbHBfhd
The same benchmark account posting its results: Jev still in the lead, but close. It is here in the Laya column because "close" is the finding that matters for an open model running on your own machine — and because the number moves the moment a fine-tune lands.
- X
Jev is cool. So is it's OSS companion, Laya. The Latest Cool Thing In AI™ tends to get a lot of hype, sometimes without everyone even understanding it. So... what is this thing? Jev is an AI model that consumes input and produces output VERY differently than chat, claude,…
A patient explainer for readers who have heard the name and cannot say what it does: how a decision model differs from a chat model, why the output has no prose to parse, and what that changes about cost. Posted by someone with no stake in either project, which is why it is near the top of this wall.
-
Jev 发布没几天,开源社区已经开始疯狂复刻了🔥 最值得推荐的五个模型: 1、Laya 421M:原生决策模型,支持 Mac 2、Decider-2B:最像 Jev,基于 Qwen3.5 3、NanoJev 0.6B:专门的 Decision Head 4、Reflex:Qwen3.5 + Direct Logits 5、System-One 4B:专门做概率校准
A Chinese-language roundup of the five open models that appeared within days of Jev, with Laya at the top of the list. It is useful as a map of the week rather than as an evaluation: five names, one line each, no measurements.
- X
Laya is the open-source Jev that already beats close Jev on latency and calibration. - 10x faster than TypeSafe Jev on the published numbers. - Laya is multilingual, and already ahead of Jev on speed. - 38ms single query vs 400ms Locally. - No per-token bill. - A System-1…
The claim-heavy version: ten times faster on published numbers, multilingual, no per-token bill, and a latency comparison of 38 ms local against 400 ms remote. Every figure comes from a benchmark somebody else ran, and the post does not say which.
-
Please welcome @Nandakishorm1, the guy behind Laya, a key open-weight, open-research predecessor to Jev and broader auto-regressive models with RL workstram!
A well-known practitioner introducing the person who built Laya, framing the work as the open-research predecessor to Jev. Worth a card for the framing alone: it is the credit argument this model's whole reception has turned on.
- X
Benchmarked Jev with Qwen3-4B and Laya (400M parameter model). Based on results here and others I've seen online, strongly suspect Jev to be in 30Bn range (compare 4bn results on MMLU vs Jev). Latency I got was 300ms, same as you'd get for a 30bn model on a good GPU. Also…
A benchmark run by someone with no stake, comparing Jev, a 4B open model and Laya on the same tasks — and then inferring that Jev is around 30B parameters from the gap. The inference is labelled as an inference, which is the part we trust.
- X
ANNOUNCING: Open Jev Playground Try all the open-source alternatives to Jev as APIs > 50% cheaper than Jev 🤯 > $0.021 per 1M tokens > Hosted as OpenAI-compatible endpoints Explore the playground ↓ https://t.co/yTcmwoN2tm
A hosted playground for the open alternatives, announced with a price and a compatibility claim: OpenAI-shaped endpoints at half the cost of the hosted model or less. It is the first post here that treats these open weights as something to resell rather than something to run.
- X
Open sourced "Jev " that runs 50x faster own Jarvis running locally now on mac 1GB RAM. 7ms per decision. laya-mlx : - 7-14ms decisions - under 1GB RAM - choice/score/yes-no, one forward pass, calibrated probabilities - no cloud, no tokens, no chat - 60+ moves/sec playing Snake…
The second post from the same account, and the more specific one: 7 to 14 ms per decision, under a gigabyte of memory, and a claim that a local assistant setup finished in a single pass. The memory ceiling is the number that decides whether this runs on the laptop you already own.
-
You may have heard of Jev, an interesting new AI. There's an open source version, Laya. I'm building Laya into Omarchy M so you can run Laya locally on the Apple GPUs and ANE. You will always have a System 1 decision model available, no subscription, no Internet connection.
A distribution argument rather than a benchmark: the open weights going into a Linux distribution's hardware integration so that a decision model is always present, offline and unbilled. If the pattern holds, this is the shape it takes when it stops being news.
-
I just discovered Laya, an open-source Jev-like model, and went down the rabbit hole. The more I look, the more Jev feels like exceptional marketing around ideas that already existed. Before Jev, open-source projects like GLiClass were already doing general-purpose zero-shot
Someone who went looking for what happened a year ago and came back with a list of prior art — GLiClass and others already doing zero-shot classification — and an argument that Jev's contribution is the framing. Disagreement with it is easy to find in the same thread on Hacker News.
- X
The problem isn't the models, it's how you define the problem space, design the movements / decisions and optimize for it (batching e.g.) Jev wins for a network-based model any day. Local models tied - Laya-mlx and my custom build Jax. Look at them go. https://t.co/fhg1vMpjzf
A practitioner's post that refuses the comparison everyone else is making: the problem is not which model is better but how you define the decision space and batch the calls, and on that reading the local model and the hosted one came out tied in his own build.
- X
Looked deeply into Laya. It's interesting. The timeline claims are fuzzy and benchmarks are mixed but the model is worth following. Laya, as compared to Jev by the author and @benchmarkheaven: -#4 overall w/ 70.1 vs Jev’s 75.4 -Much stronger cost score: 86 vs 52 -Weaker…
A careful reader's summary that refuses both camps: the timeline claims are fuzzy, the benchmarks are mixed, and the model is worth following anyway. He puts the numbers beside the cost advantage, which is the trade most of this corpus argues about without stating.
- X
🚨 THIS IS WHAT A “SECOND BRAIN” FOR AI AGENTS LOOKS LIKE. 𝗡𝗮𝗻𝗱𝗵𝗮 𝗞𝗶𝘀𝗵𝗼𝗿 built Laya — a fast, non-autoregressive decision engine designed to make decisions instead of generating long answers. Instead of asking an LLM to explain everything, Laya returns the decision…
A 'second brain for agents' explainer with the architecture animated, aimed at people building agent loops rather than at people choosing a model. The framing is promotional; the useful part is the split between deciding and explaining.
- X
Jev just got an open-source rival and it's fast as f**k Convai Innovations dropped Laya - an open-source non-autoregressive decision model that beats Jev on most benchmarks at 6-8x the speed > 421M params, ModernBERT-large backbone, Apache 2.0 > single forward pass in ~33ms on…
The licence, the parameter count, the backbone and the per-query latency, in one screen — a rare post in this corpus that puts the Apache-2.0 line and the speed claim in the same paragraph, which is what a reader deciding whether to adopt it actually needs.
- X
They Open sourced a Jev Alternative 6 to 8 times faster than Jev, Laya is a completely open, horizontal System 1 decision model family. It runs in 32.8 ms on a single GPU, with full support for over 100 languages. We're hosting it for free 🧵 https://t.co/qfvXAP8any
An offer to host the open model for free, with the three question types spelled out and a thread promising the rest. Worth a card because self-hosting is the part most readers will not do, and somebody offering to do it for them is a real answer to that.
- X
The open source model quietly destroying Jev? Laya just went head to head with a $40M funded AI startup — and the numbers are wild. @typesafeai 's Jev launched with a former OpenAI researcher, forty million dollars in funding, and a bold claim: a new category of "System 1" model…
The hype cut of the comparison, with the funding figure and the launch story attached: a $40M round, a former OpenAI researcher, and an open model going head to head with it days later. Its numbers come from the publisher's own benchmark table, which is worth holding in mind while watching.
- X
THIS IS F**KING GOLD I found Laya, which now runs natively on Apple Silicon Not via a cloud API Not via PyTorch Not by generating JSON token by token laya-mlx runs open-weight decision models locally using MLX It won't write code or a post for you But in a single pass, it…
A first-person account of getting the MLX port working on Apple silicon and being impressed by what does not happen: no cloud call, no PyTorch, no tokens generated one at a time. Enthusiastic, and specific about the install, which is why it made the wall.
- X
Laya 是 Convai Innovations 开源的编码器式决策引擎,用 choice、score、noul 三种提问原语对文本、邮件、工单或 JSON 提问,一次前向传播出结果,不做文本生成,因此没有需要解析的输出。 https://t.co/XsDEzJKv79 https://t.co/aH1ZpG67rm
A Chinese-language description of the three question types that treats them as an API rather than as a novelty, which is the right level for this model. Short, technical, and it links the model card instead of a video.
- X
💥Exploded! The open-source local version of Laya crushes Jev in response speed, with millisecond-level decisions flying by! Same Snake game. Same typed decisions. 30 seconds of testing: 🔹 Laya — 421M open-source local model → Score: 46 → Length: 52 → 86.5 decisions/sec →…
Another Snake comparison, this one with the loop's own numbers: 86.5 decisions a second, a score and a snake length recorded alongside. A game is not a benchmark, and the post does not pretend otherwise — but the rate is real and so is the recording.
- X
⚡️ 𝗟𝗮𝘆𝗮 𝗢𝗽𝗲𝗻 𝗦𝗼𝘂𝗿𝗰𝗲 𝗗𝗲𝗰𝗶𝘀𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹 𝗖𝘂𝘁𝘀 𝗜𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲 𝗟𝗮𝘁𝗲𝗻𝗰𝘆 𝘁𝗼 𝟯𝟯 𝗠𝗶𝗹𝗹𝗶𝘀𝗲𝗰𝗼𝗻𝗱𝘀 Laya answers in one forward pass through bidirectional encoders, returning a calibrated probability instead of tokens. Three primitives…
The 33 ms number with the mechanism attached: bidirectional encoders, a calibrated probability instead of tokens, three primitives. A promotional account's summary, but it states the architecture correctly, which is more than most of the summaries here manage.
-
sokudan-ja-310m: open Japanese "System One" decision model (Jev-style, no text generation). Beats laya-multilingual on all 3 primitives on a 300-item unseen-schema bench. Also: laya-multilingual never picks the first-listed score option, in ja *and* en. https://t.co/BQbpPeuWaA
A third-party Japanese model that claims to beat Laya's multilingual checkpoint on a 300-item benchmark built on an unseen schema — and adds an observation worth checking: that the checkpoint never chooses the first-listed score option, in either language.
-
@ShimazuSystems Laya is fast but it’s nowhere near as good as jev. It’s like llama 1 vs the original ChatGPT, one is a cool research project the other one is a useful tool
The dissent, kept on purpose: a reply arguing that the fast model is a research project next to the hosted one, with the comparison that one is a useful tool and the other is interesting. Two likes and no evidence, which is what it is — but a wall with no dissent on it is an advertisement.
-
Laya shipped right after Jev claiming to be faster for AI agent decisions. curious if it was accurate too I built a POC same Google ADK triage agent, same 30 labeled tickets, only the engine swapped. Jev: 422ms, 53% accuracy. Laya: 152ms 10% accuracy Faster not more accurate
The most valuable Laya post in this batch, because it reports a loss. Same agent, same thirty labelled tickets, only the engine swapped: Jev 422 ms at 53%, Laya 152 ms at 10%. Three times faster and five times worse, written as a warning.
- X
Can a classifier stop an AI coding agent from running dangerous or malicious shell commands? I tested @typesafeai's Jev against Laya, run locally on CPU and GPU. Laya on GPU: 4.5× faster. Attacks let through: 44 of 111. Jev: 6. At 856 attempts: 114 vs 9. Faster isn't safer.…
The most useful post on this wall by a distance, because it is the one that reports a loss: the same author tested both models as a shell-command safety classifier and the faster one let 44 of 111 attacks through against six. Posted as a warning, and it reads like one.
-
我做的这个测试,值得一看,因为 Laya 的出现正好回答了:能否在本地运行类似 Jev 这样的模型? 所以这个测试主要测试 Jev 和 Laya 的能力,从实测数据来看,对于同一个测试机,Jev 的判断力还是要比 Laya 更全面——这句话的意思是,Jev 问题的覆盖面更广,某些问题中 Laya 的判断力不如 Jev。 但是否
A Chinese-language test write-up with a conclusion that cuts against the enthusiasm: on the same machine Jev covered more of the questions asked of it, though the post is careful to say the gap is about coverage rather than about speed, where the local model wins.
How these were chosen
Thirty-four posts were read in full; twenty-two are published and twelve were rejected, each with the reason written down in the queue — a bare shortened link, a joke with no information in it, an automated feed account, a token launch using the model's name, and two replies that were written by a platform's own model rather than by a person. The dissenting posts are here too, including the one whose author found the faster model failing a safety test that the slower one passed.
Source: X's public syndication endpoint No API key, no OAuth, nothing rehosted The Laya topic