Check your fallback config: DeepSeek in Claude Code quietly burns tokens on V4-pro
AlchainHust · x · 2026-09-17
A practical heads-up: if you wire DeepSeek's API into Claude Code or similar agents, audit your model configs. The author found that despite switching the main model to V4.1-flash, stale fallback and subagent settings kept routing tokens to the pricier, worse-performing V4-pro, wasting money. Check main, fallback, and subagent model settings.
More from coding & agent
- Tencent unveils Hyra 2.0, a research agent that builds reusable scientific tools — TencentHunyuan · 2026-11-02
- General Reasoning explains how to train agent swarms with RL and latency pressure — rosstaylor90 · 2026-11-02
- "Coding isn't the bottleneck anymore — usage windows are," says founder Kyle Gawley — kylegawley · 2026-11-02
- Demo passed ≠ production-ready: 5 questions to answer before letting agents run your business — _kellhus · 2026-11-02
- An AI agent passed 100 test chats—still not ready to run the business — djgoel123 · 2026-11-02
- Agent given total creative freedom produces a 1:44 code-drawn animated short film — ivegotnoidea1 · 2026-11-02