1.FareedKhan-dev / kimi-k3-in-cA 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.avx2c99cpu-inferencedeep-learningfrom-scratchinference-engine+14 moreCAPACHE-2.09.9K total1.6K forks+665 starsfrom 9,202 · score 6.50
2.antirez / ds4DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCmCMIT24K total2.4K forks+106 starsfrom 24,116 · score 4.67
3.karpathy / llama2.cInference Llama 2 in one file of pure CCMIT20K total2.7K forks+30 starsfrom 20,220 · score 3.43
4.eunomia-bpf / agentsightlightweight system-level observability for AI AgentsagentebpfllmobservabilityCMIT769 total117 forks+13 starsfrom 756 · score 2.64