github nyldn/claude-octopus v9.10.2
v9.10.2 — 8-Provider Consistency + GPT-5.4 Mini

6 hours ago

8-Provider Consistency

  • embrace.sh now detects all 5 CLI providers (Codex, Gemini, Copilot, Qwen, Ollama) and dynamically builds dispatch strategies. When 3+ CLIs are available, all join the fleet.
  • Debate skill — Copilot (🟢) and Qwen (🟤) join as supplementary participants alongside the core four (Codex/Gemini/Sonnet/Opus).
  • Smart setup prompt — Detects when legacy Codex/Gemini users have new providers installed and proactively shows them.

Model Update

  • gpt-5-codex-minigpt-5.4-mini — OpenAI's latest mini model. 2x faster, uses 30% as much quota as GPT-5.4. Updated across dispatch, models catalog, provider routing, pricing, and docs.
  • All other models verified current: gpt-5.4, gemini-3.1-pro-preview, gemini-3-flash-preview, qwen3-coder, deepseek-r1-0528.

Emoji Fix

  • Qwen: 🟠→🟤 (resolved conflict with Sonnet which uses 🟠 in debates).

Full Changelog: v9.10.1...v9.10.2

Don't miss a new claude-octopus release

NewReleases is sending notifications on new releases.