Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
C
Merge pull request #1792 into integ/release-batch-5
JustVugga33d495aabbf
diff --cc c/glm53.cindex ff005627,77795a03..4aa767bd--- a/c/glm53.c+++ b/c/glm53.c@@@ -4179,9 -3927,9 +4220,10 @@@ static int serve_one(GModel *m, Tok *torss_gb(), prompt_tokens, limited);#ifdef COLI_VULKANglm53_vk_report(m, "turn");+ g53c_report(); /* the dense chain's line, when it ran */#endiffree(sequence);+ free(keys);return input_eof ? -1 : 0;}