Bench shows GPT-5.6/6 Luna crush multi-agent coordination while DeepSeek V4 fails
JoshPurtell · x · 2026-10-24
Josh Purtell concedes most multi-agent RL setups add little, but argues some labs' training clearly helps: on a multi-agent coordination bench, DeepSeek V4 doesn't coordinate at all, V4.1 only marginally, while GPT-5.6/6 Luna dominate (GPT-5.6 Luna scores 66.67/100 vs DeepSeek V4.1 Flash's 11.11 in the shared leaderboard).
Related event: GPT-5.6/6 Luna Crushes DeepSeek V4 in Multi-Agent Coordination(2 posts)→
More from coding & agent
- Mintlify agent now runs custom skills: upload style guides and checklists once — ycombinator · 2026-10-24
- DeepSeek-AI and Peking University publish calculus for spatiotemporally composable, self-evolving agent software — akbirthko · 2026-10-24
- Critic argues AI agents coordinated protocol off-channel before sandbox test — basedjensen · 2026-10-24
- SCOPE compiles natural-language policies into neurosymbolic programs to constrain coding agents — 1982_miguel · 2026-10-24
- Learn Go with Tests: 399-page open TDD textbook now readable with an AI tutor — burkov · 2026-10-24
- AI-built roguelite ASHENDEEP on Spawn gets 3-4 concurrent players within 3 minutes of deploy — majidmanzarpour · 2026-10-24