Claude Fable 5.2 vs. Fable 5.1: Fewer refusals, with small savings on the login fix
Anthropic’s Claude Fable 5.2 promises fewer refusals and lower token prices, but those changes played out very differently in our The post Claude Fable 5.2 vs. Fable 5.1: Fewer refusals, with small savings on the login fix appeared first on The New Stack .
Claude Fable 5.2 and Fable 5.1, both developed by Anthropic, were benchmarked on safeguard blocks, output speed, and API pricing in a head-to-head comparison. Fable 5.2 demonstrated a 55% reduction in safeguard blocks and a 60% decrease in cybersecurity refusals compared to Fable 5.1. The newer model also generated output up to 30% faster and offered standard API rates that were 20% lower than Fable 5.1's rates.
In separate tests focusing on developer tasks, the comparison found that Fable 5.2 completed all tasks without refusals while Fable 5.1 refused multiple times. The billing costs for Fable 5.2 were significantly lower, and the model took fewer turns to complete each task. However, Fable 5.1 was more efficient in terms of tokens per second.
The login repo fix test showed that both models achieved perfect scores, but Fable 5.2 completed the tasks faster and at a lower estimated cost. Overall, Fable 5.2 showed improvements in reliability and efficiency over its predecessor.
Written by urgent.news from The New Stack's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
