Agent Arena Pareto frontier: Claude and Kimi lead in cost-performance efficiency
arena · x · 2026-08-25
Agent Arena released the Pareto frontier based on net improvement vs. cost per task. Claude Opus 5 (High) and Claude Fable 5 (High) show the highest net improvement (12%). Kimi K3 (Max) ranks third with a 10.41% improvement rate and lower cost ($0.70). GPT 5.5 and DeepSeek V4 Pro are also listed as Pareto optimal models.
More from coding & agent
- SUCCESSOR Ω: Neural-Symbolic System Generates and Evolves Executable World Programs — Ghost_Pilot_MD · 2026-08-25
- AI Fact-Checker Audit: 1 in 18 Citations Were Fabricated — jonathancheckwise · 2026-08-25
- Guide: Running Hermes Agent on a Raspberry Pi — LeviTurk · 2026-08-25
- Vibe coding debate: Building an MMORPG solo in 2026 — TAbrodi · 2026-08-25
- Ox Alpha on track to hit 6 trillion tokens processed today — AccBalanced · 2026-08-25
- OpenAI dev livestream: driving Codex with a voice agent, keyboard optional — OpenAIDevs · 2026-08-25