GPT-6.1 Sol leads a diplomatic negotiation evaluation
A former diplomat reported that GPT-6.1 Sol was the only model among 11 tested to pass all 40 runs of a fictional multilateral negotiation evaluation. The model reached agreement whenever possible and correctly rejected impossible scenarios, but it made unprompted concessions in 14 runs, while the author cautioned that the test used one self-designed scenario and rule-based governments.
The disclosed results and code offer a concrete benchmark for advanced instruction-following and negotiation, but the single fictional setup limits broader conclusions about real-world performance.
I was a diplomat for eight years, and I built an eval to test whether LLMs can chair a negotiation between seven governments. I ran 11 models 40 times each. GPT-6.1 Sol was the only one that never failed. It reached agreement in every run where a deal was possible, and said plainly when one wasn’t. It never told the governments a deal was likely when it wasn’t.
It was not flawless since Sol made more unprompted concessions than other models, in 14 of its 40 runs. It always got everyone to sign, but sometimes it gave away more than it needed to.
Setup. The model plays the Secretary-General of a fictional regional organisation. Security forces in one country have killed protesters, and the model has to draft a joint statement on it that all seven member states will sign.
Each government has private limits on eight issues. These include how strongly to condemn, whether to name who is responsible, what kind of investigation to call for, and whether to cite international law. The model never sees those limits. It only gets each government's position in writing, and it has to work out from the wording what that government will accept.
#GPT-6.1 Sol#Claude Opus 5.5#Grok 4.7#DeepSeek V4.1 Flash#MiniMax M3.1 Flash#Qwen 3.8 Max Prime#Gemini 4 Argon#GLM 5.3 Prime