Lineup for Inside The AI Coding Stack (7/1):
@nvidia — Harry Kim on GPU infra for AI-native workloads
@FriendliAI — Gon Chun on frontier AI inference for agents
@MiniMax_AI — Victor Su-Ortiz on M3 + reasoning
@kilocode — Brian Turcotte on agentic coding in…
FriendliAI is heading to @icmlconf 2026 in Seoul this July 🇰🇷
Find us at Booth B115 — let's talk serving frontier open-weight models faster and more reliably, plus inference for coding agents.
📅 Book a 30-min meeting: calendly.com/jeongyeon-choi-…
🎟️ Side event on July…
GLM-5.2 is now available on FriendliAI — #1 output speed on OpenRouter, 99.99% uptime SLA.
Built for agents running at scale.
📌 Run GLM-5.2 now: friendli.ai/models/zai-org/G…
Link
zai-org/GLM-5.2 API & Inference Endpoint | FriendliAI
Run zai-org/GLM-5.2 with FriendliAI…
Frontier performance doesn't have to mean frontier costs.
GLM-5.2 (max) performs within 7% of Claude Opus 4.8 on agentic tasks — at less than 25% of the cost.
The intelligence your agents need. The economics your infra team won't fight you on.
Discuss this model
Add corrections, implementation notes, pricing changes, or usage caveats for other readers.