Not AI
I like big .vimrc and I cannot lie
- Sofia, Bulgaria
-
16:49
(UTC +03:00) - https://ggerganov.com
- @ggerganov
- user/ggerganov
Block or Report
Block or report ggerganov
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
1,142 contributions in the last year
Less
More
Contribution activity
April 2023
Reviewed 8 pull requests in 1 repository
ggerganov/llama.cpp
8 pull requests
- Clean up QK and file and tensor types
- Feature: Added api for getting/setting the kv_cache
- Fix memory bugs in loading code
- Use getopts for example scripts
- 10+% performance improvement of ggml_vec_dot_q4_0 on AVX2
- examples: add gpt4all script
- Benchmark test case for q4_0 matrix multiplication
- Use -march=native -mtune=native on x86 (Also Enables AVX512 on macOS)








