ngxson/llama.cpp

Pull Requests Commits

test new llama_split_prefix

ngxson committed 1 year ago

93069368

fix mmap buffer management

slaren committed 1 year ago

02020b04

llama_model_loader: fail if backend cannot allocate buffer

phymbert committed 1 year ago

d8b567d2

Handle optional tensors

phymbert committed 1 year ago

Verified 1c931f3d

Simplify this by making these optional, switch some layer creation tensor optional

phymbert committed 1 year ago

Verified c34a5dee

avoid copying the entire vector

phymbert committed 1 year ago

00381b07

llama_model_loader: PR feedbacks:

phymbert committed 1 year ago

1892ae7e

Merge branch 'master' into hp/split/load-model

phymbert committed 1 year ago

60a87ae0

ci : fix indentation error (#6195)

Vaibhavs10 committed 1 year ago

Verified 1943c019

build : add mac pre-build binaries (#6182)

Vaibhavs10 committed 1 year ago

Verified 5e43ba87

Add ability to use Q5_0, Q5_1, and IQ4_NL for quantized K cache (#6183)

ikawrakow committed 1 year ago

Verified 76aa30a2

Add nvidia and amd backends (#6157)

AidanBeltonS committed 1 year ago

Verified c5b8595e

split: move llama_tensor_offset to llama_model_loader

phymbert committed 1 year ago

18ff6ca8

Avoir copying the entire vector

phymbert committed 1 year ago

Verified b8feff41

cuda : fix conflict with std::swap (#6186)

slaren committed 1 year ago

Verified 42e21c68

split: support in llama_model_loader

phymbert committed 1 year ago

7c64fef9

cuda : print the returned error when CUDA initialization fails (#6185)

slaren committed 1 year ago

Verified 1c51f98a

llava : update MobileVLM-README.md (#6180)

ZiangWu-77 committed 1 year ago

Verified f9c7ba34

llava : add MobileVLM_V2 backup (#6175)

ZiangWu-77 committed 1 year ago

Verified 272935b2

cuda : refactor to remove global resources (#6170)

slaren committed 1 year ago

Verified ccf58aa3

Server: version bump for httplib and json (#6169)

ngxson committed 1 year ago

Verified 91f8ad16

gitignore : ignore curl-related files

ggerganov committed 1 year ago

Verified 6b7e76d2

server : allow to override -ngl in tests (#6170)

ggerganov committed 1 year ago

Verified bc0baab2

Revert "llava : add a MobileVLM_V2-1.7B backup (#6152)"

ggerganov committed 1 year ago

Verified d795988d

llava : add a MobileVLM_V2-1.7B backup (#6152)

ZiangWu-77 committed 1 year ago

Verified f8c4e745

Server: Handle n_keep parameter in the request (#6174)

jkarthic committed 1 year ago

Verified 47cc7a7b

server tests : more pythonic process management; fix bare `except:` (#6146)

cebtenzzre committed 1 year ago

Verified bd60d82d

update readme sycl for new update (#6151)

NeoZhangJianyu committed 1 year ago

Verified 6c0b2877

increase igpu cluster limit (#6159)

abhilash1910 committed 1 year ago

Verified d26e8b66

Remove undeed header file. (#6158)

dranger003 committed 1 year ago

Verified d8b009a9

Older