 Stephan Walter
|
36d19a603b
Remove Q4_3 which is no better than Q5 (#1218)
|
há 2 anos atrás |
 Georgi Gerganov
|
55390bcaf2
ggml : sync ggml (ggml_alibi)
|
há 2 anos atrás |
 0cc4m
|
7296c961d9
ggml : add CLBlast support (#1164)
|
há 2 anos atrás |
|
Georgi Gerganov
|
574406dc7e
ggml : add Q5_0 and Q5_1 quantization (#1187)
|
há 2 anos atrás |
|
Georgi Gerganov
|
7a32fcb3b2
ggml : add Q8_0 quantization format (rename the old one to Q8_1) (ARM NEON) (#1179)
|
há 2 anos atrás |
|
Georgi Gerganov
|
8a0f8673ba
ggml : export symbols (#1155)
|
há 2 anos atrás |
|
Georgi Gerganov
|
12b5900dbc
ggml : sync ggml (add GPT-NeoX RoPE implementation)
|
há 2 anos atrás |
 Kawrakow
|
38de86a711
llama : multi-threaded quantization (#1075)
|
há 2 anos atrás |
|
Georgi Gerganov
|
e0305ead3a
ggml : add Q4_3 quantization (#1082)
|
há 2 anos atrás |
 slaren
|
8944a13296
Add NVIDIA cuBLAS support (#1044)
|
há 2 anos atrás |
|
Georgi Gerganov
|
77a73403ca
ggml : add new Q4_2 quantization (ARM only) (#1046)
|
há 2 anos atrás |
|
slaren
|
315a95a4d3
Add LoRA support (#820)
|
há 2 anos atrás |
 Ivan Komarov
|
f266259ad9
Speedup the AVX-512 implementation of ggml_vec_dot_q4_0() (#933)
|
há 2 anos atrás |
|
Georgi Gerganov
|
e95b6554b4
ggml : add Q8_0 quantization for intermediate results (#951)
|
há 2 anos atrás |
 Pavol Rusnak
|
c56b715269
Expose type name from ggml (#970)
|
há 2 anos atrás |
 Kerfuffle
|
c9a59b70a5
ggml : add unary and binary map operations (#874)
|
há 2 anos atrás |
|
Georgi Gerganov
|
a3a2a0eda8
ggml : add GGML_DEFAULT_N_THREADS
|
há 2 anos atrás |
|
Stephan Walter
|
3e6e70d8e8
Add enum llama_ftype, sync ggml_type to model files (#709)
|
há 2 anos atrás |
|
Georgi Gerganov
|
c3ac702e5e
ggml : add ggml_cont() + optimize ggml_cpy() for contiguous dst
|
há 2 anos atrás |
 comex
|
f963b63afa
Rewrite loading code to try to satisfy everyone:
|
há 2 anos atrás |
 unbounded
|
62cfc54f77
Add quantize-stats command for testing quantization (#728)
|
há 2 anos atrás |
|
Georgi Gerganov
|
986b6ce9f9
ggml, llama : avoid heavy V transpose + improvements (#775)
|
há 2 anos atrás |
 Marian Cepok
|
c0bb1d3ce2
ggml : change ne to int64_t (#626)
|
há 2 anos atrás |
 Justine Tunney
|
6f23ba5ee2
Ensure --mlock works properly with mmap() support
|
há 2 anos atrás |
|
Slaren
|
c03ae8dca1
Add mmap support for model files
|
há 2 anos atrás |
|
Stephan Walter
|
c1f885067c
ggml : introduce structs for the q4 data blocks (#356)
|
há 2 anos atrás |
|
comex
|
563cdc391d
Support calling mlock() on loaded model data on Linux and macOS (#453)
|
há 2 anos atrás |
|
Stephan Walter
|
69c92298a9
Deduplicate q4 quantization functions (#383)
|
há 2 anos atrás |
|
Georgi Gerganov
|
f5a77a629b
Introduce C-style API (#370)
|
há 2 anos atrás |
 hoangmit
|
6eac39ba95
Add RMS norm and use it (#187)
|
há 2 anos atrás |
