Qwen Code SDK v0.1.12 ships microcompaction fix that preserves prompt-cache reuse
github-actions[bot] · ghdev · 2026-09-10
Qwen Code SDK TypeScript v0.1.12 (bundling CLI 0.23.3) is out:
- Managed memory availability now respects memory.enableManagedAutoMemory; hosts disabling managed auto memory no longer get remember/dream requests or managed-memory prompt instructions.
- Size-triggered microcompaction now clears old tool results to a low watermark instead of rewriting the conversation prefix each turn, preserving provider prompt-cache reuse across long tool-heavy sessions and cutting cached-input cost.
- Also includes ACP agent auth fix and parentToolCallId/subagentType fields for ACP subagent tracking.
More from coding & agent
- Stanford's PaperFormalizer turns research papers into Lean-verified versions with author in the loop — StanfordUniversity · 2026-10-27
- Hugging Face TRL makes score centering + masked importance sampling default in async GRPO — _lewtun · 2026-10-27
- AutoResearch agent's weird training curve gets dug into instead of discarded — ZimingLiu11 · 2026-10-27
- Firecrawl offers 2.5x credits to a16z Speedrun founders, unveils $75M Series B — devdigest · 2026-10-27
- Chums: open-source Apple Watch app puts Claude Code and Codex on your wrist — brandon_galang · 2026-10-27
- Open-source Clairic Studio turns pasted text into typed data and working code in 5 languages — dfinke · 2026-10-27