vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-04-28 03:30:20 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3
9773 commits 3 branches 237 tags 1 GiB
Commit graph

9773 commits

This branch
This branch
All branches
Author SHA1 Message Date
Concedo
4f8f0e5949 move embeds into their own dir, detach sd vocab into separate files 2025-10-03 14:21:09 +08:00
Concedo
c00ae93421 makefile fix vulkan noext compile (+1 squashed commits) 
Squashed commits:

[eae88fd49] makefile fix vulkan noext compile
 2025-10-02 23:19:45 +08:00
Concedo
f2ad0b78d4 refactor of added images done 2025-10-02 21:22:35 +08:00
Concedo
df87da4694 wip refactor ref image bufs 2025-10-02 16:37:17 +08:00
Concedo
539db70eac clip to cpu by default 2025-10-02 10:56:51 +08:00
Wagner Bruna
ac6be8ab8c
sd: do not force T5 on CPU anymore (#1769) 
We now have the clip_cpu config parameter for that. Todo: Will make clip cpu on by default
 2025-10-02 10:48:16 +08:00
Concedo
4587ccb71a prepare to refactor reference image 2025-10-02 10:41:29 +08:00
Concedo
e4c40405fb update lite 2025-10-01 23:23:10 +08:00
Concedo
e49ac6b120 allow clip_vision to be loaded via clip_l or clip_g param 2025-10-01 17:57:49 +08:00
Concedo
2fc31d36c0 gif mime type for animated images 2025-10-01 17:18:00 +08:00
Concedo
20c802a198 Merge branch 'upstream' into concedo_experimental 
# Conflicts:
#	.github/workflows/build.yml
#	CODEOWNERS
#	ggml/CMakeLists.txt
#	ggml/src/ggml-cpu/CMakeLists.txt
#	ggml/src/ggml-cpu/kleidiai/kleidiai.cpp
#	scripts/sync-ggml.last
#	tests/test-backend-ops.cpp
 2025-09-30 22:28:53 +08:00
Sigbjørn Skjæret
2df5bcf357
ci : disable ccache for android (#16348) 2025-09-30 15:38:01 +02:00
Georgi Gerganov
075c01567b ggml : bump version to 0.9.4 (ggml/1363) 2025-09-30 13:53:55 +03:00
Concedo
b3a0ba5e37 adjust max frames 2025-09-30 17:43:27 +08:00
Concedo
2201ddb759 fix tool builds 2025-09-30 16:29:11 +08:00
anavp-nvidia
a014310374
cuda : Enable CUDA Graph usage for Nemotron Nano v2 (NemotronH) (#16328) 
* Fix Nemotron Nano v2 9B not executing as CUDA Graph on NVIDIA GPUs

* fix to ensure test-backend-ops check passes
 2025-09-30 11:13:22 +03:00
Georgi Gerganov
35fb82497e
metal : dynamic simdgroups for MV kernels (#16340) 
* metal : dynamic simdgroups for MV kernels

* cont : minor
 2025-09-30 11:03:23 +03:00
Adrien Gallouët
3c62aed89f
common : simplify etag tracking by removing json (#16342) 
The JSON parser is temporarily kept only for backward compatibility. It
reads the etag from old .json files to prevent unnecessary re-downloads
for existing users.

This legacy code can be removed in a future version.

Signed-off-by: Adrien Gallouët <angt@huggingface.co>
 2025-09-30 10:36:33 +03:00
Charles Xu
f1eb1cb1eb
kleidiai : fix work size and threads sync for fp16 (#16246) 2025-09-30 10:07:20 +03:00
Concedo
9e4c29fda7 generate both gif and pick smaller (+1 squashed commits) 
Squashed commits:

[09122d052] generate both gif and pick the smaller one
 2025-09-30 14:42:58 +08:00
Concedo
4117542eae switch to msf gif 2025-09-30 13:56:41 +08:00
lhez
de41f2b7bf
codeowners: add codeowners for opencl backend (#16344) 2025-09-30 08:30:16 +03:00
Jeff Bolz
a74a0d69f3
tests: override test_set_rows::max_nmse_err to allow for occasional rounding differences (#16295) 
* tests: override test_set_rows::max_nmse_err to allow for occasional rounding differences

* apply similar error bounds to test_cpy
 2025-09-29 19:26:34 -05:00
Pascal
5f7e166cbf
Fix thinking blocks with quotes + add handling [THINK]...[/THINK] blocks (#16326) 
* fix: prevent reasoning blocks with quotes from being truncated

* chore: update webui build output

* feat: Improve thinking content parsing

* test: Adds ChatMessage component stories for different thinking blocks

* chore: update webui build output

* fix: ChatMessage story fix

---------

Co-authored-by: Aleksander Grygier <aleksander.grygier@gmail.com>
 2025-09-29 18:49:47 +02:00
Concedo
4f2b951547 Merge branch 'upstream' into concedo_experimental 
# Conflicts:
#	.devops/musa.Dockerfile
#	.github/workflows/build-linux-cross.yml
#	.github/workflows/build-riscv-native.yml
#	.github/workflows/build.yml
#	.github/workflows/docker.yml
#	CODEOWNERS
#	ci/run.sh
#	ggml/CMakeLists.txt
#	ggml/src/ggml-blas/CMakeLists.txt
#	ggml/src/ggml-cpu/CMakeLists.txt
#	scripts/sync-ggml.last
#	tests/test-backend-ops.cpp
#	tools/perplexity/perplexity.cpp
#	tools/server/README.md
 2025-09-30 00:36:38 +08:00
Concedo
1a1ebfc304 Merge commit '75a3a6c2cd' into concedo_experimental 
# Conflicts:
#	CMakeLists.txt
#	common/CMakeLists.txt
#	ggml/src/ggml-cuda/fattn-vec-f16.cuh
#	ggml/src/ggml-cuda/fattn.cu
#	tools/server/CMakeLists.txt
#	tools/server/README.md
 2025-09-30 00:33:36 +08:00
Concedo
4b1c89ca5c can save animated gifs 2025-09-29 22:52:42 +08:00
Georgi Gerganov
d72f5f7ba2
ci : add AMD runners and workflows (#16249)
Some checks failed
Python Type-Check / pyright type-check (push) Has been cancelled
Details 
* ci : add AMD runners and workflows

* ci : move AMD jobs to separate workflow

* cont : fix paths
 2025-09-29 17:51:48 +03:00
alex-spacemit
b77e6c18e1
ggml: riscv: add riscv spacemit backend (#15288) 
* ggml: add spacemit backend

Change-Id: I249bdc043485d815a9c351867137bc1e27cc2e23

* add new line at end of file

Change-Id: I889ed1c85fb45e62350ecde0c06f70450cadfbe2

* add riscv zba extension limit

Change-Id: I321eb200f859751727afe5cae13074dfce2bb0ce

* fixed for review comments, file renamed and format

Change-Id: Ia20b6ec24a36638e62e0fe07cf100916a7cce3ce

* fixed for code format, after clang-format

Change-Id: I5dc33a0412da3d3f2d77075d8939185d3009eca2

* use _Float16 instead of __fp16

Change-Id: I039fb02bb95270e641bc4442204e658735859d43

* add ci for riscv64-spacemit-ime-native

Change-Id: I711c1033061df1a289ea77891b2997599dfe8279

* update debian-13-riscv64-spacemit-ime-native ci label

Change-Id: Ifb2b891e2fca57b5da604fce2ac255f27731179a

* remove license comment for spacemit ime

Change-Id: If0dc3ca30a958631ccca0a28b62e0b825f9fb0c3

* upgrade binutils for gcc ime

Change-Id: Ibf2fa74c1064408974cb5b45f044d40987e5fb45

* add spacemit ime cross jobs

Change-Id: I80d74909941d41cb9cd09e51d8baf01c985cbfc6

* remove native compile for riscv64-spacemit-ime

Change-Id: I01920afafdc73fa7424014fd648d243f8ec9e25e

* ci : add caching for spacemit ime cross toolchain

Change-Id: Ic54a192019a2fd982bbd58225ce3bbc38f4053de

* ci: bug fixed for cache path and env

Change-Id: I28c42e10b6fff053bb6580926ca2353448cb042a

* Update .github/workflows/build-linux-cross.yml for cache path

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* bugfixed for  build-linux-cross.yml,  syntax error

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

---------

Co-authored-by: cailinxi <linxi.cai@spacemit.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
 2025-09-29 17:50:44 +03:00
Georgi Gerganov
2ddd3f2356 sync : ggml 2025-09-29 17:43:58 +03:00
Georgi Gerganov
4d3d455d3c sync : whisper.cpp (ggml/1359) 
* ggml : Fix MKL detection by quoting BLAS_INCLUDE_DIRS (whisper/3426)

* sync : whisper.cpp
 2025-09-29 17:43:58 +03:00
Daniel Bevenius
c9b1c06467 ggml : remove -dev suffix from release version (ggml/1355) 
This commit removes the `-dev` suffix from the version string in
CMakeLists.txt and the release script. The version will now be
just be formatted as `MAJOR.MINOR.PATCH`.
 2025-09-29 17:43:58 +03:00
Daniel Bevenius
b6ae75afb4 ggml : bump version to 0.9.3 (ggml/1353) 2025-09-29 17:43:58 +03:00
Georgi Gerganov
b6dff20e2f ggml : prepare for development of 0.9.2-dev 2025-09-29 17:43:58 +03:00
Georgi Gerganov
2db78c75e4 ggml : bump version to 0.9.1 2025-09-29 17:43:58 +03:00
Rafal Lewczuk
02463ab27b
ggml-backend : add root cause in error message if loading backend library fails (#16172) 
This PR adds additional information to an error message when loading backend library via ld_load_library() fails. This helps spotting why backend library did not load (missing library, missing dependency or unresolved symbol etc.).
 2025-09-29 13:17:09 +02:00
Sigbjørn Skjæret
adc76347d7
ggml : check cuda and metal argsort limits and add test (#16323) 
* check cuda argsort limits and add test

* add metal check
 2025-09-29 11:09:00 +02:00
Aleksander Grygier
3a2bdcda0b
Improve Mobile UI for dialogs and action dropdowns (#16222) 
* fix: Always show conversation item actions

* feat: Improve Alert Dialog and Dialog mobile UI

* feat: Add settings reset to default confirmation

* fix: Close Edit dialog on save

* chore: update webui build output

* webui: implement proper z-index system and scroll management

- Add CSS variable for centralized z-index control
- Fix dropdown positioning with Settings dialog conflicts
- Prevent external scroll interference with proper event handling
- Clean up hardcoded z-index values for maintainable architecture

* webui: ensured the settings dialog enforces dynamic viewport height on mobile while retaining existing desktop sizing overrides

* feat: Use `dvh` instead of computed px height for dialogs max height on mobile

* chore: update webui build output

* feat: Improve Settings fields UI

* chore: update webui build output

* chore: update webui build output

---------

Co-authored-by: Pascal <admin@serveurperso.com>
 2025-09-29 10:37:20 +02:00
Pascal
66bb7985c3
fix: preserved zero values in chat settings inputs and textareas by switching to nullish coalescing for field values and default placeholders (#16312) 2025-09-29 09:08:41 +02:00
Vinkal
2f61c0f5bf
llama-cli: prevent spurious assistant token (#16202) 
* tools/main: llama-cli: prevent spurious assistant token (#13402)

During prompt ingestion, prompt tokens are accepted into the sampler history (for repetition penalties). The conversation-mode path then appended `common_sampler_last(smpl)` to `assistant_ss` before any new token was sampled. At that point, "last" was a prompt-side token (e.g., an input prefix), so the assistant chat message began with an extra piece.

Fix: append to `assistant_ss` only for a newly sampled (non-EOG) token. This affects only chat message assembly (`assistant_ss` / `chat_msgs` / `common_chat_format_single`); terminal stdout is unchanged. Sampling order/logits are unchanged.

Fixes #13402.

Signed-off-by: Vinkal Chudgar <vinkal.chudgar@gmail.com>

* Update tools/main/main.cpp

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* tools/main: remove outdated comment

Signed-off-by: Vinkal Chudgar <vinkal.chudgar@gmail.com>

---------

Signed-off-by: Vinkal Chudgar <vinkal.chudgar@gmail.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
 2025-09-29 10:03:12 +03:00
ddh0
3ffd0fae47
perplexity : show more kl-divergence data (#16321) 
Adds additional percentile data for displayed in the output of `llama-perplexity --kl-divergence`:
- Added 95 percentile (mirroring existing 5 percentile)
- Added 0.1 percentile (mirroring existing 99.9 percentile)
 2025-09-29 09:30:45 +03:00
Georgi Gerganov
a4a0aa5ea2
ggml : fix dependencies for ggml_set_rows (#16318) 2025-09-29 08:41:28 +03:00
Jeff Bolz
92cd103f62
vulkan: Fix validation failure in quantized flash attention (#16292) 2025-09-29 06:50:37 +02:00
Sigbjørn Skjæret
b887d2f341
ggml : fix GGML_F32_VEC_FMA argument order in ggml_vec_mad1_f32 (#16307) 
* fix GGML_F32_VEC_FMA argument order in ggml_vec_mad1_f32

* add test that fails on simd
 2025-09-28 23:15:03 +02:00
crat0z
bd0af02fc9
common : fix reasoning before forced tool call via tool_choice = required (#16264) 
* common : fix reasoning before forced tool call via tool_choice = required

* common : improve reasoning and commentary handling when tool_choice is required

(cherry picked from commit c746984956d6882c2de73d53ae2bb3bdf889e475)

---------

Co-authored-by: Alde Rojas <hello@alde.dev>
 2025-09-28 21:13:50 +03:00
Concedo
9de04ef9c6 updated sdui, not complete 2025-09-29 00:19:18 +08:00
R0CKSTAR
d9e0e7c819
ci : fix musa docker build (#16306) 
Signed-off-by: Xiaodong Ye <yeahdongcn@gmail.com>
 2025-09-28 16:38:15 +02:00
Concedo
7b71742223 added additional toggles for SD, sdoffloadcpu , sdvaecpu and sdclipcpu 2025-09-28 22:08:29 +08:00
Aaron Teo
0124ac989f
devops: switch to using ubuntu-22.04-s390x image (#16302) 
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
 2025-09-28 19:25:58 +08:00
Imad Saddik
2811c65286
Fixed a few typos in the README of the LLaMA.cpp HTTP Server [no ci] (#16297) 2025-09-28 13:04:46 +02:00