刚刚，Anthropic 官宣了 Claude Opus 5.5，这是全新 Claude 5.5 系列的第一款模型。
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.
Claude Opus 5.5 正式发布了。 刚刚，Anthropic 官宣了 Claude Opus 5.5，这是全新 Claude 5.5 系列的第一款模型。 Anthropic 基本把上一代 Fable 级别的能力，塞进了一个更便宜的 Opus 里。 官方称，Opus 5.5 在大多数任务上的表现已经接近 Claude Fable 5.1，但运行成本比 Opus 5 低 40%。 而官方 Benchmark 的提升也很猛： Terminal-Bench 4.0：66.4% Opus 5 只有 52.3%，GPT-6 Astra 是 57.9%。 Humanity's Last Exam + Tools：67.7% Opus 5 是 63.6%，Fable 5.1 是 65.6%。 GDPval-AA v2.1：1846 Elo 超过 Fable 5.1 的 1735。 CursorBench 4.0：57.8% Opus 5 是 46.6%。 OSWorld 2.0：81.8% Opus 5 是 74.0%。 尤其是 Terminal-Bench，一代直接从 52.3% 干到 66.4%，涨了 14.1 个百分点。 当然也不是所有榜都第一。 AutomationBench 上 GPT-6 Astra 还是 41.4%，高于 Opus 5.5 的 40.0%；T
Posted by Max For AI (39.9k followers) 1 h ago · 7 likes · 1.9k views · view the original post on X. Kept by the AI Radar as Frontier models.
More AI work like this
- holy mog. GPT-6 Sol pricing is insane. — @Ananth7e
- GPT 6 family models — @HarshithLucky3
- It's double-release day! — @daniel_mac8
- 🚨 GPT-6 Sol is HALF the price of GPT-5.6 Sol. — @DanDr1s
- GPT 6 Sol pricing — @HarshithLucky3
- And it's live in the Codex CLI too! Now we wait for Tibo to work his magic 🔥 — @dedene
- GPT 6 Sol and Luna are out — @HarshithLucky3
- GPT-6 Sol and Luna are dirt cheap — @scaling01
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. 42.6k 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 18:12 UTC. Full method.