Expired65% confidenceBenchmarkExact time

The GPT-5.6 series achieves performance equal to or higher than Opus 4.8 and Fable 5 at a lower cost across multiple benchmarks.

3

Sources

65%

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

In-Depth Testing of GPT-5.6's Three Models: A Full Analysis of Soul/Terra/Luna Capabilities and Experience

bilibili声译看世界7/10/2026

Related Claims

UnverifiedThe GPT-5.6 SOUL version outperforms the benchmark model Fable 5 on multiple tests and is more competitively priced.84% similar UnverifiedGPT-6 Sol and GPT-6 Luna deliver strong performance at lower cost and have pushed the Pareto frontier on the OfficeQA Pro benchmark78% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/493183

API

curl https://kongchang.com/api/v1/knowledge/claims/493183

MCP

get_claim(id=493183)