Guide: Running local Qwen3.8-27B in DeepSeek Harness with reasoning effort config
i_wayyy_over_think · reddit · 2026-08-21
The author documents wiring a locally hosted Qwen3.8-27B into DeepSeek Harness:
- Set up vLLM via Docker+WSL2 using Codex with club-3090; too many MTP draft tokens causes crashes, set SPECN=3;
- Start with npx @deepseek-ai/dsh web, add an OpenAI-compatible provider in Settings → Models (fake API key, Base URL http://localhost:8095/v1);
- Get model IDs from the backend /v1/models; Qwen27b thinks a lot, so raise max output tokens to 64k;
- The UI can't adjust thinking effort by default — edit settings.yaml via "Open Configuration File", adding a reasoningEfforts mapping (off/low/medium/xhigh) and defaults.
More from coding & agent
- Leerob's Stanford Lecture Breaks Down Always-On Agents, Harnesses and Context Engineering — dean_rie · 2026-10-16
- Dev Praises Codex Local Computer Use: 'Every Other Harness Disappoints Me' — craigsdennis · 2026-10-16
- Decision models match LLMs on routing and safety — and are 7-8x faster — randal_olson · 2026-10-16
- Temporal Serverless Workers on Cloud Run: queue-driven autoscaling, scale to zero — rseroter · 2026-10-16
- Meta's Superintelligence Lab Runs 200+ Direct Reports—A Lesson for Flat Multi-Agent Systems — peterjliu · 2026-10-16
- LangChain CEO: the most interesting agent harness work lives at the harness-infra boundary — hwchase17 · 2026-10-16