Matheus C. Feitosa

matheus-c-feitosa

AI & ML interests

Low-bit LLM inference on CPU · quantization · runtime correctness

Recent Activity

new activity 6 days ago

prism-ml/Ternary-Bonsai-1.7B-gguf:docs: point the quickstart at PQ2_0, and at Q2_0_g64 for upstream llama.cpp

new activity 23 days ago

microsoft/bitnet-b1.58-2B-4T-gguf:llama_cpp_python: gguf_init_from_file_impl: failed to read tensor info

new activity 24 days ago

prism-ml/Ternary-Bonsai-27B-gguf:Speed when CPU-only?

View all activity

Organizations

None yet

models 0

None public yet

datasets 0

None public yet