ggml-org/llama.cpp

Pull Requests Commits

examples : add Jeopardy example (#1168)

CRD716 committed 3 years ago

Verified 5fba3c01

llama : add session file format and saved sessions in main (#1169)

ejones committed 3 years ago

Verified 1481a9cf

ggml : add helper debug printf in soft_max

ggerganov committed 3 years ago

Verified 11d90236

ggml : add CLBlast support (#1164)

0cc4m committed 3 years ago

Verified 7296c961

Correcting link to w64devkit (#1214)

Floating-Pin committed 3 years ago

Verified 78ec5437

Add Manjaro CUDA include and lib dirs to Makefile (#1212)

JohannesGaessler committed 3 years ago

Verified 92a6e13a

add avx2 for dot_q8_0_q8_0, 2x faster than scalar (#1211)

YannFollet committed 3 years ago

Verified 04aaae1d

ggml : slightly faster AVX2 implementation for Q5 (#1197)

sw committed 3 years ago

Verified 0b2da205

readme : add quantization info

ggerganov committed 3 years ago

Verified f9be42ad

ggml : add Q5_0 and Q5_1 quantization (#1187)

ggerganov committed 3 years ago

Verified 574406dc

Allow setting the rng seed after initialization. (#1184)

asgeir committed 3 years ago

Verified 87a6f846

Updating build instructions to include BLAS support (#1183)

daniandtheweb committed 3 years ago

Verified ea3ad7eb

quantize : use `map` to assign quantization type from `string` (#1191)

prusnak committed 3 years ago

Verified 859fee6d

Update SHA256SUMS after quantization change (#1181)

sw committed 3 years ago

Verified 4afcc378

py : cast lora_alpha to int in convert-lora-to-ggml (#1170)

ostix360 committed 3 years ago

Verified 667c5013

nix: use convert.py instead of legacy wrapper convert-pth-to-ggml.py (#981)

prusnak committed 3 years ago

Verified bb98e77b

ggml : add Q8_0 quantization format (rename the old one to Q8_1) (ARM NEON) (#1179)

ggerganov committed 3 years ago

Verified 7a32fcb3

ggml : use full range for Q4_0 and Q4_2 quantization (#729)

unbounded committed 3 years ago

Verified dd0eabc0

ggml : fix bug in ggml_compute_forward_sum_f32 (#1162)

xaedes committed 3 years ago

Verified 54bb60e2

ggml : export symbols (#1155)

ggerganov committed 3 years ago

Verified 8a0f8673

examples : add save_load_state example (#1150)

xaedes committed 3 years ago

Verified 0c569234

llama : increase scratch buffer size for 65B (ref #1152)

ggerganov committed 3 years ago

Verified 957c8ae2

examples/main README improvements and some light refactoring (#1131)

mgroeber9110 committed 3 years ago

Verified 9b0a4d42

Fix build for gcc 8 and test in CI (#1154)

sw committed 3 years ago

Verified 2ec83428

Fix cuda compilation (#1128)

slaren committed 3 years ago

Verified e4cf982e

llama : refactor get / set state + remove redundant kv cache API (#1143)

ggerganov committed 3 years ago

Verified c4fe84fb

Fix LoRA acronym (#1145)

slaren committed 3 years ago

Verified 1d78fecd

scripts : add helper scripts to synch ggml repo

ggerganov committed 3 years ago

Verified 284685f1

Added README.md for main with examples and explanations (#1139)

DannyDaemonic committed 3 years ago

Verified edce63ba

ggml : do not print perf ops that have not been used at all

ggerganov committed 3 years ago

Verified ec9cdb67

Newer Older