Unverified50% confidenceOpinionExact time
The poster speculated that GPT-6 and GPT-6.1 Sol share the same pre-training base but adopt different post-training schemes and differ in the number of recurrent passes.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/6/2026
First Seen
Valid until: 1/4/2027
Sources
Related Entities
Related Claims
UnverifiedGPT-5.6 Sol能够自主完成对Luna的后训练，既能扮演策略模型执行训练，也能扮演奖励模型提供反馈信号80% similarVerifiedGPT-5.6 Sol achieves state of the art on the coding benchmark Terminal Bench.76% similarVerifiedGPT-5.6 Sol has autonomously completed post-training for Luna.76% similarUnverifiedGPT 系列、Llama 等模型的核心能力绝大部分来自预训练而非后续微调阶段76% similarUnverifiedGPT-5.6 has been specifically trained to autonomously determine when tasks should be delegated to subagents.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/980685API
curl https://kongchang.com/api/v1/knowledge/claims/980685MCP
get_claim(id=980685)