Perplexity: Mixed-Model Agent Teams Score Higher and Cost Less
perplexity_ai · x · 2026-11-18
Perplexity shares results on mixed-model agent teams: on Terminal-Bench 4.1 and WANDR, every mixed team beat four copies of its lead model, gaining +0.008 to +0.125 and +0.015 to +0.064 respectively.
Costs favor teams too — the best mixed teams cost $10.31 and $54.50 per run versus $13 and $66.41 for four lead-model copies. On WANDR, three of four teams also finished sooner than the model working alone: teams are faster, better, and cheaper.
Related event: Perplexity Finds Mixed-Model Agent Teams Beat Identical Model Copies(4 posts)→
More from coding & agent
- Fable 5.2 turns a fake mobile game ad into a real playable game: 70 min, 2,982 lines, $29.52 — Dan_Jeffries1 · 2026-11-18
- Google-Backed Kaggle Agent Contest Offers $400K to Crack Engineering Benchmark PLE — amankhan · 2026-11-18
- Dev snark: before running 10 agents, ask if you could even build software with one terminal — KevinNaughtonJr · 2026-11-18
- Fan-built zero-socket MCP bridge gives Project Zomboid an AI radio Game Master — Triminum · 2026-11-18
- Suno confirms its MCP includes visuals generation — suno · 2026-11-18
- Oxlint adopts tsrs backend: type-aware linting up to 7x faster, 75% less memory — cnakazawa · 2026-11-18