cd /news/large-language-models/fable-5-median-thinking-declined-in-… · home topics large-language-models article
[ARTICLE · art-136085] src=twitter.com ↗ pub= topic=large-language-models verified=true sentiment=↓ negative

Fable 5 – Median thinking declined in August

Anthropic's Fable 5 model delivered dramatically fewer thinking tokens in August than in July after the company made the model permanently available in subscription plans, according to measurements by Lon Lundgren. Lundgren reported that reasoning declined over the entire period and fluctuated across multi-day episodes, with most invocations receiving little to no thinking tokens even at xhigh or max effort levels, and longer thinking runs almost never reaching published benchmark levels. Lundgren advised users to "Ask about the inference regime you were served" rather than assuming the model was nerfed.

read2 min views1 publishedSep 21, 2026
Fable 5 – Median thinking declined in August
Image: source

Lon Lundgren on X: "After Anthropic made Fable 5 permanently available in subscription plans, I noticed a large drop in performance. The model felt dumber, and I couldn't explain why. Measured five different ways, August delivered dramatically fewer thinking tokens than July." / X

Lon Lundgren on X: "After Anthropic made Fable 5 permanently available in subscription plans, I noticed a large drop in performance. The model felt dumber, and I couldn't explain why. Measured five different ways, August delivered dramatically fewer thinking tokens than July."

After Anthropic made Fable 5 permanently available in subscription plans, I noticed a large drop in performance. The model felt dumber, and I couldn't explain why. Measured five different ways, August delivered dramatically fewer thinking tokens than July.

After Anthropic made Fable 5 permanently available in subscription plans, I noticed a large drop in performance. The model felt dumber, and I couldn't explain why. Measured five different ways, August delivered dramatically fewer thinking tokens than July.

This wasn't a one-time drop. Reasoning fell over the entire period, and fluctuated across multi-day episodes. Some of these fluctuations aligned with specific product announcements and releases. I started to see how the model could feel great one day, and terrible the next.

I was consistently using an xhigh or max effot level, but when I looked deeper, I found that most invocations to the model were receiving little to no thinking tokens at all. And when longer thinking runs did happen, they almost never reached published benchmark levels.

This unfolded across a six week data capture and analysis odyssey, and led to a number of surprising findings. Next time the model feels dumber, don't ask if the model was "nerfed". Ask about the inference regime you were served, instead.

Buddy its so simple. If one person used fable. It'll blow your socks off. Divided by millions It starts to suck. Models get dumber as usage goes up until they make another datacenter. Then on and on forever

── more in #large-language-models 4 stories · sorted by recency
── more on @anthropic 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/fable-5-median-think…] indexed:0 read:2min 2026-09-21 ·