Running remote Ollama+Qwen3.8 on RunPod via Codex with just an API key
huangyun_122 · x · 2026-10-29
huangyun122 demos an "A2A"-style workflow: from local Codex, spin up Ollama + Qwen3.8 on a remote RunPod machine by providing just an API Key — no manual Jupyter setup needed. Once configured, Codex can spawn subprocesses to call Claude, Qwen, and other heterogeneous LLMs and switch between them.
The screenshots show Codex autonomously picking Qwen3.8-27B Q80 (30GB), wiring it into local OpenCode and pi clients with medium thinking mode, downloading the 29GB weights in shards, and after stalled shard connections switching to a faster path: downloading on the local Windows machine then SSH-transferring the remaining shards back to the Pod.
The author muses that with enough budget, a regular home Windows box could instantly orchestrate Mac Minis placed around the world through Codex/Claude.
More from coding & agent
- Dev's code merged by AI agent with zero human intervention — iamsahaj_xyz · 2026-10-29
- Turning a laggy old Xiaomi phone into an offline MP3 player with Claude — PeanutGrinderZ_2091 · 2026-10-29
- 10 open-source repos that give coding agents a code map, led by 125k-star Graphify — victor_explore · 2026-10-29
- Snaptrude launches AI agent that takes building designs from site to construction-ready LOD 400 models — EXM7777 · 2026-10-29
- DecisionRAG replaces the retriever with a decision model, beating 8B rerankers by 13.8 R@1 — omarsar0 · 2026-10-29
- Turning photos, charts and logos into animated videos in 5 minutes with Claude Code and Remotion — TawohAwa · 2026-10-29