ngxson/llama.cpp

Pull Requests Commits

do not activate -cnv for non-instruct models

ngxson committed 1 year ago

73c4e77d

update readme (2)

ngxson committed 1 year ago

dd221bdd

update readme (writing with the help of chatgpt)

ngxson committed 1 year ago

723d77ce

Merge branch 'master' into xsn/cli_auto_cnv

ngxson committed 1 year ago

e6e9a6f5

contrib : add naming guidelines (cont) (#11177)

ggerganov committed 1 year ago

Verified a29f0870

server : (UI) Support for RTL text as models input or output (#11208)

ebraminio committed 1 year ago

Verified 437e05f7

contrib : add naming guidelines (cont) (#11177)

ggerganov committed 1 year ago

Verified ca001f66

common : support tag-based --hf-repo like on ollama (#11195)

ngxson committed 1 year ago

Verified 00b4c3da

add warn on bad template

ngxson committed 1 year ago

5b1f7108

contrib : add naming guidelines (#11177)

ggerganov committed 1 year ago

Verified 7426a26b

llama : remove 'd' from bad special token log (#11212)

danbev committed 1 year ago

Verified 8f70fc3d

cli : auto activate conversation mode if chat template is detected

ngxson committed 1 year ago

9d7d5f21

ggml : do not define GGML_USE_CUDA when building with GGML_BACKEND_DL (#11211)

rgerganov committed 1 year ago

Verified 1244cdcf

Reset color before we exit (#11205)

ericcurtin committed 1 year ago

Verified 924518e2

llama : fix chat template gguf key (#11201)

ngxson committed 1 year ago

Verified 9a483999

llama : remove notion of CLS token (#11064)

ggerganov committed 1 year ago

Verified 08f10f69

llama : add `llama_vocab`, functions -> methods, naming (#11110)

ggerganov committed 1 year ago

Verified afa8a9ec

gguf-py: fixed local detection of gguf package (#11180)

VJHack committed 1 year ago

Verified c05e8c99

convert : sort print supported models [no ci] (#11179)

danbev committed 1 year ago

Verified 2739a71e

examples : add README.md to tts example [no ci] (#11155)

danbev committed 1 year ago

Verified ba8a1f9c

convert : add --print-supported-models option (#11172)

danbev committed 1 year ago

Verified ff3fcabc

Vulkan: Fix float16 use on devices without float16 support + fix subgroup_size_control validation error (#11161)

0cc4m committed 1 year ago

Verified c3f9d257

llama: add support for QRWKV6 model architecture (#11001)

MollySophia committed 1 year ago

Verified ee7136c6

SYCL: Refactor ggml_sycl_compute_forward (#11121)

qnixsynapse committed 1 year ago

Verified c6860cc7

doc: add cuda guide for fedora (#11135)

teihome committed 1 year ago

Verified 1204f972

server : add tooltips to settings and themes btn (#11154)

danbev committed 1 year ago

Verified 8eceb888

model: Add support for PhiMoE arch (#11003)

phymbert committed 1 year ago

Verified f8feb4b0

media : remove old img [no ci]

ggerganov committed 1 year ago

Verified be0e950c

llama-chat : add phi 4 template (#11148)

ngxson committed 1 year ago

Verified d9feae1c

fix: add missing msg in static_assert (#11143)

hydai committed 1 year ago

Verified 8d59d911

Older