给大家带来claude opus 5.5 的前端测试结果. 直接说结论, 我怀疑现在opus 5.5就是 fable 5.1 的量化版或者自家蒸馏版, 可以直接看我的视频,…
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.
给大家带来claude opus 5.5 的前端测试结果. 直接说结论, 我怀疑现在opus 5.5就是 fable 5.1 的量化版或者自家蒸馏版, 可以直接看我的视频, 几乎看不出两个模型实现细节上的差别. 而且模型继承了Anthropic一贯的优良特点, 一个字, 稳, 6次抽卡6次全都是这个质量. 除此之外我测试时很明显 opus 5.5 的思考token消耗更多. 这意味着opus5.5模型会更小一些(用reasoning长度换性能). 这同样意味着会有雷霆大思考的情况, 比如最后一个火山喷发测试, 我terminal和网页都测试了好几次, 结果都无法输出代码. 思考卡住了. 当然瑕不掩瑜. 毕竟比fable便宜又能获得差不多的性能, 只要不降智, 就是好模型(但无奈Anthropic降智是祖传艺能...). > #opus55 #claudeopus55 #anthropic
Posted by karminski-牙医 (39.6k followers) 1 h ago · 4 likes · 1.3k views · view the original post on X. Kept by the AI Radar as Frontier models.
More AI work like this
- That’s exactly why I’m confused. MANY of my sources are saying GPT-6 Sol and Luna are… — @TokenGremlin
- Luna is 100x cheaper than Astra. — @0x0SojalSec
- We had early access to GPT 6 Sol and Luna and ran them on Ramp Accounting Bench. — @RampLabs
- Opus 5.5 is great at frontend! And funny enough performs better without a design skill! — @daradoescode
- Opus 5.5 vs GPT 6 astra — @notjazii
- Anthropic’s new Opus 5.5 beats Fable 5.1 on AAII at ~2.5x lower cost. — @cline
- when you’re an AI maxi and GPT-6 SOL + Opus 5.5 launch on the same day — @kloss_xyz
- The top model on @OpenRouter the last 30 days just got better & 50% cheaper🐺🌔 — @nickbaumann_
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. 43k posts from 5k X accounts over the last 14 days, 1.8k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-22 20:11 UTC. Full method.