The Price of Thought: A Multilingual Analysis of Reasoning, Performance, and Cost of Negotiation in Large Language Models
This work addresses the challenge of AI negotiation performance and cost trade-offs, with incremental insights into multilingual reasoning consistency.
The study tackled the problem of evaluating how reasoning affects negotiation abilities in large language models across three languages, finding that enabling reasoning improves GPT-5's performance by 31.4% but increases cost by nearly 400%, and reveals a multilingual distinction where open-weight models switch to English for reasoning while commercial models maintain language consistency.
Negotiation is a fundamental challenge for AI agents, as it requires an ability to reason strategically, model opponents, and balance cooperation with competition. We conduct the first comprehensive study systematically evaluating the effect of (LLM-)reasoning on the negotiation abilities of both commercial and open-weight LLMs, and do this across three languages. Using a self-play setup across three diverse dialogue games, we analyse trade-offs between performance and cost, the language consistency of reasoning processes, and the nature of strategic adaptation exhibited by models. Our findings show that enabling reasoning-that is, scaling test time compute-significantly improves negotiation outcomes by enhancing collaboration and helping models overcome task complexities, but comes at a substantial computational cost: reasoning improves GPT-5's performance by 31.4 % while increasing its cost by nearly 400 %. Most critically, we uncover a significant multilingual reasoning distinction: open-weight models consistently switch to English for their internal reasoning steps, even when negotiating in German or Italian (and thus possibly impacting potential explainability gains through the disclosure of reasoning traces), while leading commercial models maintain language consistency between their reasoning and final output.