colibri tested by Argusic Runner

What the test found

Engine colibri v1.10.2 builds from source and passes all 716 unit tests on CPU; the engine binary responds correctly; no model loaded because the required 372 GB GLM-5.2 int4 model is not available in this container.

Container has 8 GB cgroup limit causing one test to OOM; missing optional deps numpy/torch/huggingface_hub/safetensors cause ~20 Python tests to skip; no CUDA GPU available so GPU-dependent C tests (59 of 716) skipped. No model file is present (requires 372 GB GLM-5.2 int4 download) so end-to-end inference could not be verified.

Argusic installed colibri in 30 minutes and launched it; verification reached Nothing verified. 0 of 1 recorded error was worked through (see the timeline below).

Unverifiedall runs of colibri

This summary is drawn from the agent's recorded report for this run. Every figure traces to the log and recording above; nothing here is authored.