Have been using Fable 5 all day just continuing what I was doing with Opus
The findings are true
It's completely nerfed
Both of them, in their own way, are correct.
The short version: The model didn't get dumber. The gatekeeper in front of it got much more aggressive. That distinction matters a lot depending on what you use Fable for.
What BridgeBench actually measuredBridgeMind—an AI evaluation platform—re-ran its full coding suite against the July 1 version of Fable 5 the day it came back.
FABLE 5 CAME BACK NERFED.
We re-ran the July 1st version of Claude Fable 5 on BridgeBench.
The results are brutal:
Debugging: 86.2 → 25.9Refactoring: 73.6 → 38.4Hallucination: 75.9 → 61.7
The catch is in the methodology. Of 12 TypeScript debugging tasks, only three actually reached Fable 5. The remaining nine were intercepted by Anthropic's new safety classifier and rerouted to Claude Opus 4.8—and BridgeBench scores every fallback as zero, because the model that answered wasn't the one under evaluation.
What Arena.AI actually measuredThe community has been asking how Claude Fable 5 compares before vs. after its latest re-deployment.
We collected thousands of votes on the new endpoint across Arenas - Text, Vision, Document, Code, and Agent - and here’s an early score preview.
In other words, when Fable 5 actually handles the task, it still performs like Fable 5. The frustration on X isn't about a worse model but more about paying for a model that often isn't the one answering.
Who's affected, who isn'tGeneral users doing creative writing, document analysis, research, and expert-level text queries will likely notice little to no difference. Those are the categories where Arena.AI shows flat or improved performance. If there is some improvement, it might be too small to notice, especially in subjective, qualitative tasks like creative writing, where it is hard to fully measure results.
So, basically, writers, researchers, and analysts will get the Fable 5 they expected. Developers are a different story.
Anyone working in security-adjacent territory—coding memory management, anything touching words like "vulnerability," "exploit," "hook," or even "fix"—is going to hit the fallback regularly.
The gap between BridgeBench's collapse and Arena's stability comes down to task type. BridgeBench loads its suite with exactly the kind of code-repair and debugging prompts that trigger the new classifier. Arena's human voters ask a much wider mix of things, and most of them don't look like exploit code to a safety layer.
Anthropic has given no target date for when that will happen.
