Event ·Multi-source

Bindu Reddy 评测：Fable 5.2 表现不佳，被称为 flop

Bindu Reddy 在 X 上发文称 Fable 5.2 是个 token 消耗大户，性能却没有达到预期，直言其 flops。她认为 Anthropic 目前最好的模型是 Opus 5.5 和 Fable 5.5，而非 5.2 版本。文中还提到 Carbon 在 Fable 5.2 上的采用率正在上升，但整体评价仍是负面。

Current read

AirOps 前 CEO Bindu Reddy 直接点名 Fable 5.2 吃 token 又不出效果，还把 Anthropic 自家的 Opus 5.5 拉出来对比。如果你在选模型做生产部署，这篇第一手评价可以帮你避坑，省下测试成本。

11 sources33° AI heatLast updated 2026/11/15 06:41:35

Evidence

11 sources

Primary sourceBindu Reddy

2026/11/15 06:41:35View source

Related source 1lmarena.ai

2026/11/13 20:03:24View source

Related source 2Decoder

2026/11/14 16:48:01View source

Related source 3Shubham Saboo

2026/11/13 06:44:37View source

Related source 4Garry Tan

2026/11/14 04:45:28View source

Related source 5kimmonismus

2026/11/13 16:35:20View source

Related source 6Andrew Curran

2026/11/13 20:22:26View source

Related source 7小互

2026/11/14 04:54:46View source

Related source 8向阳乔木

2026/11/14 08:10:36View source

Related source 9Thomas Wolf

2026/11/14 15:34:12View source

Related source 10shao__meng

2026/11/15 09:35:27View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story