ggml-org/llama.cpp

Pull Requests Commits

Apply min_p to unsorted tokens (#5115)

JohannesGaessler committed 1 year ago

Verified 9241c3a2

Tests for min_p, sampling queue (#5147)

JohannesGaessler committed 1 year ago

Verified b2b2bf98

readme : add link to rust bindings (#5148)

MarcusDunn committed 1 year ago

Verified af4980bf

llama : add support for Orion-14B (#5118)

sharpHL committed 1 year ago

Verified f2e69d28

docker : add server-first container images (#5157)

K-Mistele committed 1 year ago

Verified 39baaf55

llava : support for Yi-VL and fix for mobileVLM (#5093)

cmp-nct committed 1 year ago

Verified 6db2b41a

ggerganov committed 1 year ago

Verified 753eafed

ggml : check ggml_add src1 type (ggml/708)

foldl committed 1 year ago

Verified e9764230

Remove unused data and add fixes (#5154)

MKlimenko committed 1 year ago

Verified 35a2ee91

server : add self-extend support (#5104)

Maximilian-Winter committed 1 year ago

Verified ec903c03

Add OpenCL add kernel (#5151)

0cc4m committed 1 year ago

Verified a1d6df12

cmake : pass CPU architecture flags to nvcc (#5146)

cebtenzzre committed 1 year ago

Verified bbe7c56c

cuda : fix tensor size calculation for non-split buffer (#5145)

slaren committed 1 year ago

Verified 62fead3e

ggml-alloc : add 10% margin to the buffer sizes (#5149)

slaren committed 1 year ago

Verified 15b4538f

ggml : update softmax n_task calculation (#5126)

snadampal committed 1 year ago

Verified 7032f4f6

scripts : move run-with-preset.py from root to scripts folder

ggerganov committed 1 year ago

Verified 5f1925a8

tests : gitignore test-c.o

ggerganov committed 1 year ago

Verified 3b7c914d

server : refactored the task processing logic (#5065)

ngxson committed 1 year ago

Verified 48c857aa

ci : add model tests + script wrapper (#4586)

crasm committed 1 year ago

Verified 413e7b05

metal : remove unused `n_buffers` and `buffers` (#5129)

ptsochantaris committed 1 year ago

Verified 6dd3c28c

gguf : fix "general.alignment" type in gguf_reader.py (#5136)

snowyu committed 1 year ago

Verified 38b431de

readme : update hot topics

ggerganov committed 1 year ago

Verified aad0b01d

Another bucket sort (#5109)

ikawrakow committed 1 year ago

Verified 1182cf4d

readme : add MobileVLM 1.7B/3B to the supported models list (#5107)

XiaotaoChen committed 1 year ago

Verified fe54033b

llama : dynamic temperature sampling (#4972)

l3utterfly committed 1 year ago

Verified 5eaf9964

examples : make pydantic scripts pass mypy and support py3.8 (#5099)

cebtenzzre committed 1 year ago

Verified d292f4f2

android : use release cmake build type by default (#5123)

ValleZ committed 1 year ago

Verified 256d1bb0

Fix Q3_K_XS for MoE models (#5113)

ikawrakow committed 1 year ago

Verified faa3526a

metal : show compile log messages

ggerganov committed 1 year ago

Verified ddc5a503

cuda : fix 2-bit quants on amd hip (#5105)

Engininja2 committed 1 year ago

Verified cd4fddb2

Newer Older