Tags

Tags give the ability to mark specific points in history as being important

b1109

95b6e521 · added `struct` to llama_dump_timing_info_yaml's `llama_context` (#2857) · Aug 29, 2023
b1108

44c117f4 · train : mem usage and other improvements (#2439) · Aug 28, 2023
b1107

43033b7b · llama-bench : set locale to utf8 (#2832) · Aug 28, 2023
b1106

6b73ef12 · YAML result logging + preset script (#2657) · Aug 28, 2023
b1105

75fafcbc · make : fix tests build (#2855) · Aug 28, 2023
b1104

be475f60 · llama.cpp : fix wrong vsnprintf call in MS compiler (#2856) · Aug 28, 2023
b1103

3af6b863 · ggml : tiny ggml_vec_dot_q4_K_q8_K AVX2 improvement (#2819) · Aug 28, 2023
ci_cublas_linux-b1104-cec628e

cec628e7 · temporarily disable broken 512 build · Aug 28, 2023
b1102

35feac65 · ggml : sync (mem align to header + conv_transpose_2d fixes + ggml_alloc) (#2852) · Aug 28, 2023
b1101

92b1bbd2 · CUDA: fix RoPE asserts, block sizes (#2833) · Aug 28, 2023
b1100

dd0dc366 · llama.h : add missing struct keyword for C compat in callback type (#2847) · Aug 28, 2023
b1099

f55538c3 · metal : fix memory leak (#2762) · Aug 28, 2023
b1098

ebcee207 · quantize : make output filename optional again (#2823) · Aug 28, 2023
b1096

103cfafc · gguf : fix strings to not be null-terminated (#2839) · Aug 27, 2023
b1095

c10704d0 · llama : fix MPI threads (close #2827) · Aug 27, 2023
b1094

230d46c7 · examples : update llama2.c converter to read vocab and write models in GGUF format (#2751) · Aug 27, 2023
b1093

463173a6 · llama : speedup tokenization (#2831) · Aug 27, 2023
b1092

eaa13a48 · falcon : fix CUDA inference by making K and Q contiguous (#2830) · Aug 27, 2023
b1089

a6d1189f · k_quants tuning for Falcon-7b (#2816) · Aug 27, 2023
b1087

d0cee0d3 · gguf : add 64-bit support (GGUF v2) (#2821) · Aug 27, 2023

Prev
1
…
60
61
62
63
64
65
66
67
68
…
99
Next