由 @StepFun_ai Step 5 Preview 已经上线，表现十分惊人，展现出了在 Real API Pricing Pareto 占据帕累托前沿的潜力
This is a AI post classified by Jev as Frontier models (a model release), kept by the AI Radar because it carries real work, not commentary.
由 @StepFun_ai Step 5 Preview 已经上线，表现十分惊人，展现出了在 Real API Pricing Pareto 占据帕累托前沿的潜力 Step 5 Preview 在AA Index 榜单得分44分，与 Kimi K3(max) 和 Grok 4.6(xhigh)处于同一水准； 在 Terminal-Bench 4.0中得分33.3%，与 Muse Spark1.3相同，同时十分接近 GPT 5.6 Terra 的 35.4%； 并且，这只是一个 600B / 活跃 27B MoE 的模型，总参数和 DeepSeek V4.1 Flash 差不多！ 💰Step 5 Preview 定价 $0.05/$1.00/$2.70，输入（缓存命中）/输入/输出 （每 Mtokens)，按real api pricing 统一负载均衡算为 $0.0797/MTok (缓存97.5% / 输入 2.15% / 输出 0.35%)，在AA 的 Index Cost 中每任务成本仅为 $0.71，仅为同44分数的 Kimi K3 （$2.00)和 Grok 4.6($2.32) 的约三分之一 如果在 StepPlan中提供的额度保持不变或不削减太多，那Step 5 将以非常低的价格实现接近前沿级别水准,目前我在测试一些有趣的东西，很期待看到效果!
Posted by FeiZ (8.1k followers) 2 h ago · 16 likes · 1.5k views · view the original post on X. Kept by the AI Radar as Frontier models.
More AI work like this
- Step 5 Preview: Coding Strength, Uneven Capabilities — @ZhihuFrontier
- @Luna11054 — @Luna11054
- Fable 5.2 is pure JS thats it one shot — @chetaslua
- 🚨 Fable 5.2 Leak: Beats GPT-6 Astra — @Priyannkaaaa
- Best way to test Step-5-preview is with comparison. — @ItsmeAjayKV
- Step 5 Preview — @HarshithLucky3
- 🚨 Step 5 Preview is actually pretty cracked — @LuminaBench
- 🚨 StepFun just dropped Step 5 Preview — @LuminaBench
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 34.2k posts from 4.8k X accounts over the last 14 days, 1.4k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-20 10:08 UTC. Full method.