Not Diamond releases model routing method, cuts costs 20-80%
rohanpaul_ai · x · 2026-09-02
Not Diamond released their model routing methodology, achieving Opus-level high quality while cutting agent costs by 20-80%. Addressing the difficulty of routing for long-running agents, they treat routing as a sequential decision problem, predicting future reward and cost using session state, KV-cache, and task complexity. This achieves Pareto-dominance on interactive benchmarks.
Related event: Not Diamond Unveils Agent Model Routing Method, Cutting Costs Up to 80%(2 posts)→
More from coding & agent
- Slash Claude costs: How to save 15M tokens monthly with /doctor and config tweaks — dotey · 2026-09-02
- Meta paper: Agents complete tasks but fail to prevent catastrophic actions like factory resets — rohanpaul_ai · 2026-09-02
- Hamel Discusses Avoiding Overfitting in AI Skill Optimization — HamelHusain · 2026-09-02
- AnySearch: RL Framework Adapts Single Search Policy to Any Budget — _reachsumit · 2026-09-02
- FreshCtx 0.9.0 fixes Agent TOCTOU gap by validating evidence freshness — Street-Chest2270 · 2026-09-02
- Dense Process Supervision for Search Agents via Fact Utility Estimation — _reachsumit · 2026-09-02