opencode v2 bug: warming prompts sent during long requests kill local sessions
arren36 · ghdev · 2026-10-18
A reproduced issue in anomalyco/opencode (#56636): after upgrading v1→v2, a long local CPU-only session reloading 200K context scheduled a duplicate warming task alongside the main request. With timeouts disabled, the warming request was retried 3 times at 15-minute intervals, then the main context reload was stopped 5 minutes after the last retry, killing the session. The author notes docs don't state warming should be disabled for local models, only that the default interval is 4 minutes (user had set 10 min).
More from coding & agent
- Claude cut a 2.6-hour video export to 20 minutes — but only after being told to rebuild the server — byhewar · 2026-10-23
- 3 flight-booking rules my AI agent checks that I'd never do by hand — Efistoffeles · 2026-10-23
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23
- How do you audit AI agent decisions in regulated workflows? A field survey — Ok_Way_1993 · 2026-10-23