@10xmylife: Tried it out today, and it really works, so we'll go with this plan.
Summary
A user tested the Jev model's performance in the game 'Slay the Spire 2', where its decision-making speed was extremely fast, taking only 0.7 seconds, far surpassing humans and GPT-6 Astra, which is astonishing.
View Cached Full Text
Cached at: 09/19/26, 07:08 PM
Tried it out today, and it really works, so we’ll go with this plan.
paulwei (@coolish): Yesterday’s release https://t.co/6eiMFaJTrj
The Jev model is designed for high-frequency decision-making — does that mean it’s perfect for gaming too?Previously used GPT-6 Astra to play Slay the Spire 2
Capable but slow.Just tested Jev in action: decision and execution took only 0.7 seconds.
I could barely see the screen before it had already made its move.
This superhuman game speed truly left me jaw-dropped again 😅Hard to put into words — watch the video below ⬇️
#AIinGaming #JevModel
Similar Articles
@10xmylife: cool 啊
An AI system named 'jev' is showcased playing Super Smash Bros. against itself, controlling multiple characters and making rapid decisions within fractions of a second, using over 22 million tokens.
@reach_vb: 6.1 Terra beats Astra at vision tasks all whilst being ~47% faster and ~92% cheaper
据 SkalskiP 的测试，GPT-6.1 Terra 在文本、目标检测和计数等视觉基准上超越 Astra，同时速度快约 47%、成本低约 92%，结果令人难以置信。
jevman: AI decision models play Pac-Man
作者将六个流行的决策模型（jev 1.13、Kev 4B、Clef、Clef Flash、GPT-6 Luna 和 Laya）放在《吃豆人》游戏中进行实时对决，jev 1.13 以平均 2,750 分领跑排行榜。团队已公开排行榜和开源仓库，任何人提交自己的模型即可参赛。
@AravSrinivas: Ultrafast inference of OpenAI frontier models now supported on Computer. Try it out. It’s super fast. Credit drain warn…
Perplexity 宣布在其 Computer 产品中支持 OpenAI 前沿模型（GPT-6 Astra、SOL 6.1）的超高速推理模式，供 Max 订阅用户使用，可更快完成长任务，但会消耗更多积分。
@RookieRicardoR: Domestic models break through again, matching top models like Claude 4.6 and Gemini 3.1 Pro. Just tested Qwen3.7-Max, sharing some real thoughts. Last night I topped up as soon as the API went live and chose three tasks (see video) to test Qwen3.7-Max's frontend capabilities…
The user tested Qwen3.7-Max and believes it matches top models like Claude 4.6 and Gemini 3.1 Pro in frontend, computing power, and Agent capabilities. Its reasoning ability has significantly improved, and with monthly iteration speed, it has become a first-tier domestic model.