Argus multi-model code review

The registry has 15 reviewers. Three default-roster ones are dual-route (a direct provider API plus OpenRouter) and are reordered by the routing preference (the custom-only hermes-4.3 is dual-route too); the rest are single-route API reviewers or paid-CLI reviewers. Model versions are the headline of this release: GLM‑5.2, MiniMax M3, and the new default DeepSeek V4 Pro.

ReviewerRoute(s)Notes
glm-5.2 dual-routez.ai direct + OpenRouter (z-ai/glm-5.2)Strong security + logic.
minimax-m3 dual-routeMiniMax direct + OpenRouter (minimax/minimax-m3)High precision.
kimi-k2.6OpenRouter (moonshotai/kimi-k2.5, fallback kimi-k2-thinking)Long-context, agentic.
mimo-v2-proOpenRouter (xiaomi/mimo-v2-pro)1M context window.
qwen-3.6-plusOpenRouter (qwen/qwen3.6-plus)1M ctx, conservative / high precision.
grok-4.20OpenRouter (x-ai/grok-4.20)2M ctx, pricey.
deepseek-v4-pro dual-route default DeepSeekDeepSeek direct + OpenRouter (deepseek/deepseek-v4-pro)1.6T MoE, 49B active, ~1M ctx; reasoning + security.
deepseek-v3.2OpenRoutercustom-only superseded by v4-pro.
hermes-4.3 dual-routeNous direct + OpenRouter fallbackcustom-only
geminiGemini CLI (paid sub)disabled Windows .cmd tree-kill re-test pending; use gemini-or meanwhile.
gemini-orOpenRouter (google/gemini-2.5-flash)~2s/call, best value.
codexCodex CLI (paid sub, GPT-5.x)Thorough, slow; OpenRouter fallback.
claudeClaude CLI (paid sub)Auto-added to profile rosters when host ≠ claude.
opencodeOpenCode CLI (paid sub)Top benchmark performer, slow cold start.
copilot-gpt5GitHub Copilot CLIdisabled / eliminated agent harness returns prose, not our JSON schema.
🔄
Dual-route vs CLI

The dual-route reviewers respect route_preference. CLI reviewers (Codex / Claude / OpenCode / Gemini) keep their subscription primary with OpenRouter as a true fallback, no matter the routing preference. See Routing preference.

Ready-made groupings of these reviewers live in Profiles.