Cline (@cline) says Meituan LongCat-2.0 is free inside Cline now: a 1.6T open-weights MoE with 1M context that it says scores near Claude Opus 4.7 and Gemini 3.1 Pro—install the CLI and pick it under free models via /model.
Command Code (@CommandCodeAI) says Meituan LongCat-2.0 is free on all plans: a 1.6T MoE with ~48B active params and 1M context - install via npm i -g command-code. Meituan account confirmed it is live and free in Command Code.
Cline announced that GLM-5.3 Flash, previously previewed as Ox Alpha, is free in Cline. The 320B-parameter mixture-of-experts model with 18B active parameters is beating frontier models on benchmarks and has become Cline fastest-growing model, driving over 11% of all traffic in less than a week.
Cline said about half of Qwen3.8-Max usage on the product already flows through ClinePass, which it prices at roughly 5x cheaper than the direct API. For developers who standardized on Qwen3.8-Max inside the coding agent, the routing choice now dominates unit economics. Cline is pushing ClinePass as the default path, splitting model selection from the billing lane so an open-source agent can cut long-session cost without leaving the model.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.