Corridor's Argos 1: SFT'd Qwen3.5-9B beats GPT-6 and Opus 5.5 on agent security-guidance selection
LangChain · x · 2026-10-22
- Corridor released Argos 1, its first security model, built by post-training on agent trajectories from LangSmith using smithtune; they SFT'd Qwen3.5-9B.
- On selecting relevant security guidance for coding agents, Argos 1 beats GPT-6, Claude Opus 5.5, and Gemini 3.6 Flash on both quality (Micro F1 0.65 per the scatter chart) and response time.
- LangChain frames this as a great way to build alpha into agents, and is bullish on teams using their own data to drive model-layer improvements.
Related event: Corridor's 9B safety model Argos 1 beats GPT-6 and Claude Opus 5.5(3 posts)→
More from coding & agent
- GMG Open-Sources an Apache 2.0 ServiceNow Agent With Human-Approved Changes — adnan_hashmi · 2026-10-22
- RAX drives robots with a frozen coding agent; DeepSeek version beats π0.5 at 1/20 cost — du_yilun · 2026-10-22
- Stripe partners with Anthropic to run business payments and refunds inside Claude Duet — jeff_weinstein · 2026-10-22
- Aaron Levie: autonomous domain-specific agents are AI's emerging form factor, a leap past the LLM-wrapper era — _AustinCalvert_ · 2026-10-22
- Claude Duet opens as research preview: a proactive agent that watched Slack and ordered books — altryne · 2026-10-22
- Vibe-coded ComfyUI node unlocks 2MP for Fibo 2.0 Edit, built in 10 minutes with a free LLM — WhiteZero · 2026-10-22