Unverified50% confidenceBenchmarkExact time
在Agent能力测试LexExam中,GPT-5.6 Soul比Anthropic的Claude Fable 5高出13.1分
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
Valid until: 10/10/2026
Sources
GPT-5.6发布:ChatGPT合并Codex,进化为AI工作台
bilibili水山AI漫谈7/10/2026
Related Claims
UnverifiedGPT-5.6 Soul 和 Soul Ultra 的基准测试成绩大多超过了 Mysus,Terra 超越了 Fable 5,而 Luna 的表现不如上一代 GPT-5.576% similarUnverifiedGPT-5.6 SOUL版本多项测试表现优于对标模型Fable 5,定价更具竞争力74% similarUnverified从早期跑分看,GPT-5.6 Soul整体仍略低于Fable 5,但差距极小74% similarUnverified在Terminal Bench 2.1上,GPT-5.6 Soul Ultra以91.9分排名第一,Claude Mythos 5位居第三73% similarUnverified在Zapier的Automation Bench上,GPT-5.6 Soul在max档以0.7分微弱优势领先Fable,成本相近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)