19 LLMs play 376 rounds of a Codenames-style game: guessing matters more than clueing
Hardicecream17 · reddit · 2026-10-31
The developer of Word Spies, a Codenames-inspired party game, pitted 19 LLMs against each other across 376 games. Key findings: Gemini and GPT models clue and guess best (85-93% good guesses), Claude models clued surprisingly poorly; the better-guessing team won 68% of games vs 53% for better-clueing teams; ambitious multi-word clues lost badly (38% win rate over 287 games); and going first conferred no advantage. Clue grading used embeddings plus weaker LLM judges (Llama 4 Maverick, Qwen3-VL), so the author trusts win rates over grades. Sample sizes varied (18-152 games per model) and reasoning settings differed between groups.
More from Models
- Users complain Opus 5.5 feels heavily nerfed — RileyRalmuto · 2026-10-31
- Bug bounty researcher flags AI security platform sessions as potential training data of unpatched exploits — npinto · 2026-10-31
- MOSS AMA Roadmap: Full-Duplex Live Model, Image Turbo, Consumer-GPU Text2Vid — tomByrer · 2026-10-31
- Grok says rumored Gemini 4 Carbon cyber model is wrapped in guardrails, not leaked carelessly — AccBalanced · 2026-10-31
- Alleged 8TB "Claude Fable 5" weight leak from AWS S3 looks fabricated, teardown finds — AccBalanced · 2026-10-31
- NVIDIA ships NVFP4 DeepSeek-V4.1-Flash, 81GB smaller with near-identical quality — AccBalanced · 2026-10-31