6 commits

Author	SHA1	Message	Date
Justine Tunney	fe044e22cc	Switch public headers to getopt_long() entirely ... Cosmopolitan's getopt() is now redefined as __getopt().	2023-07-02 19:57:43 -07:00
Justine Tunney	0409096658	Get us closer to building busybox ... This change undefines __linux__ and adds APIs like clock_settime(). The gosh darned getopt_long() API has been reintroduced, thanks to OpenBSD.	2023-06-18 04:13:45 -07:00
Justine Tunney	e7eb0b3070	Make more ML improvements ... - Fix UX issues with llama.com - Do housekeeping on libm code - Add more vectorization to GGML - Get GGJT quantizer programs working well - Have the quantizer keep the output layer as f16c - Prefetching improves performance 15% if you use fewer threads	2023-05-16 08:07:23 -07:00
Justine Tunney	cc1732bc42	Make AARCH64 harder, better, faster, stronger ... - Perform some housekeeping on scalar math function code - Import ARM's Optimized Routines for SIMD string processing - Upgrade to latest Chromium zlib and enable more SIMD optimizations	2023-05-15 02:15:34 -07:00
Justine Tunney	282dd8e7b7	Get radpajama to build ... make -j8 o//third_party/radpajama/radpajama.com make -j8 o//third_party/radpajama/radpajama-chat.com This change gets the radpajama.mk config working. This package depends on THIRD_PARTY_GGML but it's configured to call ggjt_v1(), so that the library will provide the old quantizers. The ggml_quantize_chunk() API will now dispatch to older quantizers based on the configured version.	2023-05-13 20:44:36 -07:00
Ariel Núñez	b3e3359d22	Import radpajama (a redpajama.cpp fork) (#814) ... This is the relevant commit: bfa6466199 Model download links: https://huggingface.co/ceonlabs/radpajama/tree/main	2023-05-11 07:12:08 -07:00