OpenCode bug: reasoning chunk after finish_reason fails otherwise-complete steps
adgeese · ghdev · 2026-10-18
OpenCode issue #56455 (reproduced) reports that some OpenAI-compatible hosts send an extra chunk carrying only reasoning content after the chunk with finishreason. The OpenAI Chat parser then raises "received content after the finish reason"; since the error isn't classified as incomplete-stream, it isn't retried and the whole step fails even though the answer and tool calls were already complete. Seen on 0.1-0.2% of requests via OpenRouter with deepseek/deepseek-v4.1-flash (DeepInfra/StreamLake hosts), failing subagent sessions. It's a follow-up to the unfixed #43040; present in v2.0.32 and on dev.
More from coding & agent
- Claude Code Pushes Urgent Security Update Across All Installations — umeshonai · 2026-10-23
- 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