gpt4all

major_new_ui_redesign

fix-localdocs-stale

fix-cuda-arch-default

pull-2403

newui-refactor-mysettings

main

nightly-offline-build-4-ui-redesign

fix-cur-gpu-device

fix-moc-build-failure

macos/cloud-signing-workflow

fix-backend-includes

win-suppress-dll-errors

fix-win-unicode-libpath

v281-dev-console

cuda-early-alloc

fix-embed-over-512

fix-embedding-after-cuda

v280-release-notes

localdocs_changes

new_icon_set

major_new_ui_redesign_draft

gpt4all-2.8.0-pre1

fix-metal-build

fix-missing-batch-free

fix-archless-gguf-crash-uniqueptr

shutdown_embedding_thread

fix-reload-progress

new-chat-fixes

improve-chat-ctxmenus

fix-send-while-responding

fix-generated-name

fix-win-icon

add-cuda-support-wip

remove-docker-server

readme-update

release-notes-sign-up

release-275

mixpanel-device-stats

fix-opt-out

fix-nomic-embed-error

fix-localdocs-startup-event

v274-release-notes

fix-msvc-cpuid

add-llama3-instruct

dependabot/go_modules/gpt4all-bindings/golang/golang.org/x/net-0.23.0

feat/remove-town-hall

localdocs-fixes

temp-revert-new-ui

fix-codeblock-trimming

gpt4all-2024-roadmap

py-suppress-gpu-stderr

linux-273-debug

localdocs_contextlinks

embed4all-dynamic

docs/roadmap-update

py-listgpu

ui_changes

pull-2045

fix-py39-py310-typeddict

fix-unicode-paths

fix-readme-quants

py-fix-partial

fix_server_colors

ui_redesign

python-doc-updates

intel-mac-test

split_main

fix_rmodel_convert

restrict_chat_width

rework_chat_panel

fix_2105

manyoso-patch-6

fix_stale_model_settings

load-checkpoint

fix_2092

necessary_sort

batch_updatedata

fix_clones_2087

model-warnings

fix_for_2080

bindings-expose-fakereply

fix-resetcontext-ub

fix-linux-downloadbtn

fix_another_crasher

ts-async-stream

ci-separate-holds

pr-1897

build-chat-271

add-gpu-model-arches

fix_chatgpt_crash

fix-mistral-fname

update-modelsjson

add-models3

chatml-fix

add-gemma

save_window_geometry

model_loading_revamp

gpu-diag

network_check

pr-1915

limit-localdocs-exts

remove_docx

enable-pascal-gpus

cfg-gpu-layers

update-llamacpp-vulkan

new_ui_themes

py-bindings-readme

ci-confchange-runall

update-llama.cpp

jacoobes-patch-1

fix/macm1ts

configurable-ctx

fix-direct-avx2-link

fix-deserialize-assert

update-models-list-v260

network_retry

manyoso-patch-5

manyoso-patch-4

localdocs_fix

localdocs_v2

feat-ts/streaming

building-qt-add-libs

fix-old-lib-refs

ggufv3

more-words

remove-old-chat

v2.5.1

readme-gguf-note

py-default-ext

dll-namepat

executable-scripts

fix-main-qml

replace-ggml-refs

dll-load

gguf-python

v2.5.0

fix-model-urls

replit-1.5

llama-log-errors

remove-star-hist

py-improvements

update-py-bindings

py-win-mingw-path

kp-logger-fix

mmcleanup

fix-autoconfig

py-macos-version

no-mv-only-mm

miniorca3b-up

more-mm

replace-pkg-resources

embedding-default-model-fix

quiet-by-default

gguf-mm

always_save_chats

restore_state_from_text

change-issue-template

quiet-codespell

clearer-fallback-msg

apage43-patch-1

gguf_latest_llama

cpu-fallback-reason

atreat_latest_kernel_refactor

vulkan_bert

matmatwip

circleci-installers

gguf-mac-build-fix

vulkan_subgroups

deverbosify

fixvulkanwinmsvc

dynlog

actual_device

dupe-gpu-fix

dupe-gpu-name-fix

vkpy2

pybump-vkagain

py-vklinking

use-vk-nodynamic

m3zh-patch-1

vulkan_backend

vkwinfix

python-cleanup-3

crosspath

niansa-patch-7

fix/tstests

feat(typescript)/dynamic-template

font-sizing

evaluate

llamabump714

scrollbar-fix

light-mode-fix

unicode-decoding

pretty-jazz-62

tgi-oss

fixstarcoderjson

light-mode-chat

gpt4all-api-monitoring

fix-batched

AndriyMulyar-patch-12

rguo123/embed4all-js-fork

quantize

AndriyMulyar-patch-11

starcoder

bert_fixes

bert_latest

more-highlighting-rules

modelclone

system_prompts

subdirmodels

prefer7b

busy_for_modelsjson

7b_preferred

avxonlyapi

openai-fix

cmake-exportbuild

threadcount

mpt

json-highlighting

kvcswap

backend_impl_cleanup

pycontextfix

per_model

settings_dialog_redesign

go-bash

dlopen_gpu

python-bindings-bugfix-2

python-bindings-bugfix

modelmem2

python-bindings-love-broken

modelmem

replit-formatsync

java-highlighting

dlopen_gpu_rebase

triton-inference

consolidate_settings

settings_refactor

falcon-mem

forcemetal

mem-req-calc

modellist

mainline-llama-up

niansa-patch-6

niansa_replit_warn_fix

threads

more_fixes

fixes

typescript

token_speed

models_update

deprecated

rguo123/no-stdio

mb-mib

code-style-unification

kquant-fix

niansa-patch-5

niansa-patch-4

llmodel_better_promptfnc

rguo123/metalreplit-update

metalreplit

llmodel-shared-toktoidsv

llama-mainline-up

rguo123/pypi-ver-bump

rguo123/windows-debug

recalc_bos

fix_codespell

prompt_syntax

cpp_syntax

syntax_highlighting

refactor_context_links

metal-wip

niansa-patch-3

kquant_llama_fix

always_sync

llmodel_c_test

niansa-patch-2

update_llama_cpp

circleci_for_gpt4all

rguo123/update-models-json

circleci_for_gpt4all_v2

python_bugfix_download_hf

AndriyMulyar-patch-10

revert-chatgpt-context-bug

localdocs-label

manyoso-patch-3

release_notes_2.4.5

rguo123/update-pypi

AndriyMulyar-patch-9

remove_older_models

manyoso-patch-2

minimum_hardware

backend_prompt_dedup

junior

niansa-patch-1

rguo123/python-bindings-ggmlver-update

modelsjson-spellfix

recalcuatecontext_nonvirtual

localdocs_servermode

dlopen_backend_5

dlopen_backend_4

dlopen_backend_3

revert-747-dlopen_better_implementation_management

fix_warnings

dlopen_backend_2

hotfix/python-download-model-bug

adamt_localdocs

dlopen_backend

fix_build

settings_dialog_fixes

ui_tweaks

AndriyMulyar-patch-8

Yuvanesh-ux-patch-1

zanussbaum-patch-1

dedup_qml

fix_folderdialog

fixtab_borders

mlock_true_apple

mlock_true

AndriyMulyar-patch-7

rguo123/gpt4all-wiki

rguo123-close-issues-patch-2

rguo123/python-streaming

rguo123/small-doc-fixes

chat-doc-fixes

AndriyMulyar-patch-6

AndriyMulyar-docs-typo

fix_installers

AndriyMulyar-gpt4all-chat-docs

chatclient_docs

adamt_misc

AndriyMulyar-patch-5

rguo123/pr_template_fix_2

AndriyMulyar-patch-4

AndriyMulyar-patch-3

AndriyMulyar-patch-2

AndriyMulyar-patch-1

server_lifetime_mgmt

threaded_memory_mgmt

rguo123-readme-patch-1

modal_labs_python_docs

update_readme

httpserver

rguo123/mpt-python-bindings

manyoso_cleanup_chatllm

clear-cloudfront-cache

rguo123/docs-cicd

manyoso-patch-1

rguo123/pr-template-fix

fix_mpt_ggml

readme_update

pythia

duplicates

mosaic

accel_eval

license

gptj_eval

roadmap

train

eval

v2.8.0

python-v2.6.0

v2.7.5

v2.7.4

python-v2.5.1

python-v2.5.0

python-v2.4.0

python-v2.3.2

python-v2.3.1

python-v2.3.0

python-v2.2.1.post1

python-v2.2.1

python-v2.2.0

python-v2.1.0

python-v2.0.2

python-v2.0.1

python-v2.0.0

python-v2.0.0rc2

python-v2.0.0rc1

v2.7.3

v2.7.2

v2.7.1

v2.7.0

v2.5.1

v2.5.0

v2.6.2

v2.6.0

v2.5.4

v2.5.3

v2.5.2

v2.4.19

v2.4.18

v2.4.17

v2.4.16

python-v1.0.11

python-v1.0.12

v2.5.0-pre1

v2.6.1

v2.8.0-pre1

Commit Graph

Author	SHA1	Message	Date
Jared Van Bortel	d2a99d9bc6	support the llama.cpp CUDA backend (#2310 ) * rebase onto llama.cpp commit ggerganov/llama.cpp@d46dbc76f * support for CUDA backend (enabled by default) * partial support for Occam's Vulkan backend (disabled by default) * partial support for HIP/ROCm backend (disabled by default) * sync llama.cpp.cmake with upstream llama.cpp CMakeLists.txt * changes to GPT4All backend, bindings, and chat UI to handle choice of llama.cpp backend (Kompute or CUDA) * ship CUDA runtime with installed version * make device selection in the UI on macOS actually do something * model whitelist: remove dbrx, mamba, persimmon, plamo; add internlm and starcoder2 Signed-off-by: Jared Van Bortel <jared@nomic.ai>	4 weeks ago
Adam Treat	987546c63b	Nomic vulkan backend licensed under the Software for Open Models License (SOM), version 1.0.	9 months ago

Author

SHA1

Message

Date

Jared Van Bortel

d2a99d9bc6

support the llama.cpp CUDA backend (#2310 )

* rebase onto llama.cpp commit ggerganov/llama.cpp@d46dbc76f
* support for CUDA backend (enabled by default)
* partial support for Occam's Vulkan backend (disabled by default)
* partial support for HIP/ROCm backend (disabled by default)
* sync llama.cpp.cmake with upstream llama.cpp CMakeLists.txt
* changes to GPT4All backend, bindings, and chat UI to handle choice of llama.cpp backend (Kompute or CUDA)
* ship CUDA runtime with installed version
* make device selection in the UI on macOS actually do something
* model whitelist: remove dbrx, mamba, persimmon, plamo; add internlm and starcoder2

Signed-off-by: Jared Van Bortel <jared@nomic.ai>

Adam Treat

987546c63b

Nomic vulkan backend licensed under the Software for Open Models License (SOM), version 1.0.

2 Commits (main)