dsh-qwen38-compaction-fix
Compaction fix for NInfer-hosted qwen3.8-27b gateways: disables thinking on dsh context-compaction and session-title calls so xhigh reasoning stops burning the whole output budget; the same idea applies to other launch methods, this package targets NInfer only.
针对 NInfer 托管的 qwen3.8-27b 网关的压缩修复：在 dsh 上下文压缩与会话标题调用中关闭思考，避免 xhigh 推理耗尽全部输出预算；同一思路适用于其他启动方式，本包仅面向 NInfer。
How to install
dsh plugin add dsh-qwen38-ninfer-compaction-fix About
Qwen3.8 (NInfer) Compaction Fix Fix for a local **qwen3.8-27b** gateway running on the **NInfer** engine that fails to compact. The model thinks at xhigh, spends the entire output token budget on reasoning before reaching a conclusion, and the compaction checkpoint comes back truncated. This DeepSeek Harness (dsh) plugin disables thinking **for compaction calls only** and applies, to those calls, the sampling parameters the model recommends when thinking is off. **Scope: NInfer only.** This package is written for gateways served by the NInfer engine (ninfer-serve, OpenAI-compatible API): the w…
Recommendation signals
Meta
- License
- MIT
- Language
- JavaScript
- GitHub stars
- 2
- mo. downloads
- 208
- Last push
- 2026-09-06
- Created
- 2026-08-25
Basic safety check
- Findings
- None
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- dsh-plugin
Related plugins
dsh-plugin-subscriptions
V1ki/dsh-plugin-subscriptions
Use ChatGPT (Codex), Claude, and Grok subscriptions as DeepSeek Harness LLM providers, with Settings login, model catalogs, usage, plus image_generate, video_generate, and x_search tools.
dsh-workbuddy-connect
corrinehu/dsh-workbuddy-connect
Brings the models in the WorkBuddy desktop app straight into DeepSeek Harness — zero configuration in the DSH chat.
claude-code-router
musistudio/claude-code-router
One local control plane for every AI agent: route across models, fuse new capabilities, orchestrate tools, and stay fully in control.
dsh-thinking-effort
hytime/dsh-thinking-effort
Configure reasoning-effort levels and sub-agent defaults for custom DSH models.