要么在最便宜的模型里做到最厉害，要么就直接做最前沿的模型，否则就完全没有讨论度和使用度。
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.
感觉现在发模型只能走两个极端： 要么在最便宜的模型里做到最厉害，要么就直接做最前沿的模型，否则就完全没有讨论度和使用度。 昨晚的两个发布：Grok 4.7 和 Mimo V2.6。两者的能力在各种第三方测评上都差不多，但在价格上差别太大了。 Mimo V2.6 Pro 在缓存命中时，输入价格是 ¥0.025，输出是 ¥6。而 Grok 4.7 从美元换算过来的话，输入已经到了 ¥3.35，输出更是高达 ¥40。 看了一些早上 Twitter 上大家对 Grok4.7 的反馈和测评，主要是一个吐槽贵，另一个是觉得它效果可能没那么好，又慢又不好。 Mimo 这个价格比当前 DeepSeek V4（不管是 Flash 还是 Pro）的波谷价格都要便宜一倍以上。 比如 Mimo V2.6 Flash 输出价格是 ¥2，而 DeepSeek V4.1 Flash 输出价格是 ¥4。 另外，Mimo V2.6 目前在 OpenCode 好像下周开始限免。 Mimo也推出了自己的桌面端，似乎是单独的订阅价格，可以去试试；桌面端里面带了图像生成模型，可以配合着画一些界面之类的。 估计两家都是觉得 OpenAI 可能今天下午就会发新模型（可能是 GPT-6 Sol 和 GPT-6 Luna，好像非常便宜），想赶在 OpenAI 之前发出来，结果凑到一块儿了。
Posted by 歸藏(guizang.ai) (175.2k followers) 1 h ago · 34 likes · 9.9k views · view the original post on X. Kept by the AI Radar as Frontier models.
More AI work like this
- Announced today :) — @Biggest
- Here is the news we have all been waiting for — @ItsmeAjayKV
- Qwen 4 models have been officaly announced at the Apsara Conference. — @AiBattle_
- Xiamo launched the new AI model, which comes close to GPT 5.6 and costs a fraction of… — @Vvikramai
- Grok TRIPLED on Terminal-Bench 4.0 in only two months (12.4% -> 38.0%), overtaking… — @ns123abc
- 卧槽，Grok 4.7 怎么给我一种汪峰的感觉🤣 — @xueyu1125
- "Grok 4.7 will be the 2.1T model released a few weeks later. This will be better than… — @theo
- the hype for Chinese open models once again beats the hype for US AI Labs, MiMo-V2.6-Pro… — @TimJayas
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. 40.9k posts from 5k X accounts over the last 14 days, 1.7k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-22 04:17 UTC. Full method.