Expired50% confidenceBenchmarkExact time
In the agent capability benchmark LexExam, GPT-5.6 Soul scored 13.1 points higher than Anthropic's Claude Fable 5.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/10/2026(expired)
Sources
GPT-5.6 Released: ChatGPT Merges with Codex, Evolving into an AI Workbench
bilibili水山AI漫谈7/10/2026
Related Claims
UnverifiedGPT-5.6 Soul and Soul Ultra's benchmark scores mostly surpassed Mysus, Terra surpassed Fable 5, while Luna's performance fell short of the previous generation GPT-5.5.76% similarUnverifiedThe GPT-5.6 SOUL version outperforms the benchmark model Fable 5 on multiple tests and is more competitively priced.74% similarUnverifiedBased on early benchmark results, GPT-5.6 Soul overall remains slightly below Fable 5, but the gap is extremely small.74% similarUnverifiedOn Terminal Bench 2.1, GPT-5.6 Soul Ultra ranks first with a score of 91.9, while Claude Mythos 5 ranks third.73% similarUnverifiedOn Zapier's Automation Bench, GPT-5.6 Soul leads Fable by a narrow margin of 0.7 points at the max tier, with comparable costs.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/493644API
curl https://kongchang.com/api/v1/knowledge/claims/493644MCP
get_claim(id=493644)