vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-10-21 13:51:29 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
16423 commits 3 branches 258 tags 1,017 MiB
Commit graph

16423 commits

This branch
This branch
All branches
Author SHA1 Message Date
Concedo
76e77f3405 updated readme 2026-10-18 17:34:17 +08:00
Concedo
7f035b72ff updated colab 2026-10-18 17:13:20 +08:00
Concedo
63634ec66a updated readme 2026-10-18 17:07:10 +08:00
Concedo
f41f621e09 updated readme 2026-10-17 18:34:27 +08:00
Concedo
9be8c136ae Merge commit '23b0202a18' into concedo_experimental 
# Conflicts:
#	ggml/src/CMakeLists.txt
#	ggml/src/ggml-cpu/CMakeLists.txt
#	tests/test-batch-alloc.cpp
#	tools/server/README.md
 2026-10-16 00:15:23 +08:00
Concedo
e19a6813a9 up ver 2026-10-15 21:27:04 +08:00
Concedo
34a22215ac revert old behavior for macos 2026-10-15 12:30:15 +08:00
Concedo
6d8d651eba updated lite (+1 squashed commits) 
Squashed commits:

[b851f8b70] updated lite
 2026-10-15 00:12:50 +08:00
Concedo
c368a80fe8 show runner hardware 2026-10-14 22:25:09 +08:00
Concedo
0770b28ab7 Revert "revert https://github.com/leejet/stable-diffusion.cpp/pull/1987 due to h3 incoherence" 
This reverts commit 534925b92b.
 2026-10-14 13:24:21 +08:00
Wagner Bruna
96600d1ef7
sd: fix h3 incoherence with flash attention (#2535) 
ggml/src/ggml-cpu/ops.cpp triggered the issue when built with -Ofast .
 2026-10-14 13:24:00 +08:00
Concedo
5f70d40c87 link to koboldai lite beta site 2026-10-14 12:00:18 +08:00
Concedo
ddae02976d sort models in returned list 2026-10-14 00:24:04 +08:00
Wagner Bruna
aa2bc070c9
sd: add a non-default target for a Vulkan sdmain binary (#2534) 2026-10-13 20:16:10 +08:00
Concedo
e1c41e17ee use shared timeout for reqtimeout, sanitize values 2026-10-12 23:24:50 +08:00
Concedo
d644efd779 change some phrasings 2026-10-12 23:05:24 +08:00
Concedo
534925b92b revert https://github.com/leejet/stable-diffusion.cpp/pull/1987 due to h3 incoherence 2026-10-12 18:52:49 +08:00
Wagner Bruna
2698d1dca0
sd: model loading refactors (#2530) 
* sd: rename internal t5xxl parameter

* sd: perform all model path changes inside kcpp_set_model_paths

* sd: simplify TAE file selection
 2026-10-12 18:42:43 +08:00
Wagner Bruna
e93465c423
sd: fix VAE tiling adjustment to never reduce tile size (#2531) 
This was added back in 9450834335 for optimizing tile sizes: a
chosen high tiling threshold means the user can generate images of
that size, so e.g. a 1536 threshold shouldn't need to drop all the
way to (then) 512x512 tiles. But retrieving the default tile size
got lost in the original code, so the calculated tile size was
always applied, even for tiny values.
 2026-10-12 18:10:47 +08:00
Concedo
c3b9e61a55 updated readme 2026-10-12 14:44:25 +08:00
Concedo
8bb561d758 change text for agent 2026-10-11 21:26:22 +08:00
Concedo
768f5ab1dc harden auto mode for agent 2026-10-11 21:05:37 +08:00
Concedo
67f1315833 updated lite 2026-10-11 17:52:52 +08:00
Concedo
6e3d00e827 expose original path as well in case some tensors are referenced by original path name 2026-10-11 17:49:55 +08:00
Wagner Bruna
a84d884ca1
sd: move model path fixups from sdcpp to sdtype_adapter (#2527) 2026-10-11 17:40:31 +08:00
Concedo
37d1fb0caf updated docs 2026-10-11 17:27:54 +08:00
Concedo
8e1fb4904b wip improving docs 2026-10-11 13:44:55 +08:00
Concedo
41a86b2cf1 updated sdui 2026-10-11 11:59:18 +08:00
Pascal
23b0202a18
server: leave a busy slot untouched when a request pins it (#30295) 
A request asking for a busy id_slot still ran the prompt cache update
on that slot before being deferred. When the RAM cache held a better
match, it was loaded into the slot while another request was still
generating there, and that generation continued on the wrong context.

The busy slot is now returned as is and the request waits for it.
 2026-10-10 21:25:38 +02:00
tc-mb
69f201a205
model : support MiniCPM-V 4.7 (#29416) 
* mtmd : add MiniCPM-V 4.7 support

Signed-off-by: tc-mb <tianchi_cai@icloud.com>

* model : allow mrope time from an extra position slot

Signed-off-by: tc-mb <tianchi_cai@icloud.com>

* Update conversion/minicpm.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>

* Slim down comments

Signed-off-by: tc-mb <tianchi_cai@icloud.com>

* fix for "do not hand-wrap comments"

Signed-off-by: tc-mb <tianchi_cai@icloud.com>

* fix ci

Signed-off-by: tc-mb <tianchi_cai@icloud.com>

* rm 3d repo for pr one

Signed-off-by: tc-mb <tianchi_cai@icloud.com>

* gguf: add rope.section_order metadata

* fix comments

* handle grid layout

* allow compat

---------

Signed-off-by: tc-mb <tianchi_cai@icloud.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
 2026-10-10 20:41:59 +02:00
Aaron Teo
abee0c8476
cmake(s390x): disable z17 target for unsupported compilers (#30297) 2026-10-10 20:00:56 +02:00
Concedo
4fd0ceee30 Merge branch 'upstream' into concedo_experimental 
# Conflicts:
#	common/chat.cpp
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	ggml/src/ggml-opencl/kernels/cpy.cl
#	ggml/src/ggml-opencl/kernels/gemv_noshuffle_q4_0_f32.cl
#	ggml/src/ggml-opencl/kernels/gemv_noshuffle_q4_k_f32.cl
#	ggml/src/ggml-sycl/convert.cpp
#	ggml/src/ggml-sycl/dequantize.hpp
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	ggml/src/ggml-sycl/mmvq.cpp
#	ggml/src/ggml-sycl/quants.hpp
#	ggml/src/ggml-sycl/vecdotq.hpp
#	scripts/sync_vendor.py
#	tests/CMakeLists.txt
#	tests/peg-parser/tests.h
#	tests/test-backend-ops.cpp
#	tests/test-chat-auto-parser.cpp
#	tests/test-chat.cpp
#	tests/test-peg-parser.cpp
#	tools/llama-bench/README.md
#	tools/llama-bench/llama-bench.cpp
#	tools/server/README.md
 2026-10-10 23:40:34 +08:00
Concedo
983a696078 added /paste mode for agent 2026-10-10 22:49:25 +08:00
Concedo
e2011b5709 protection from multiline paste 2026-10-10 22:29:04 +08:00
Mikolaj Kucharski
aa94f20861
args: add LLAMA_ARG_SLOT_SAVE_PATH env for --slot-save-path (#30272) 
Allow configuring --slot-save-path via LLAMA_ARG_SLOT_SAVE_PATH so
it can be set from an EnvironmentFile in a systemd unit file.
 2026-10-10 15:34:40 +02:00
Concedo
c299b64024 auto compaction for agent 2026-10-10 19:46:23 +08:00
Concedo
9e258d9059 remove deprecated field for stream_layers 2026-10-10 18:58:54 +08:00
Concedo
d49dc15939 allow loading agent session directly from kcpp, allow setting reasoning effort 2026-10-10 18:45:13 +08:00
Concedo
7ec2115ddf removed --multiuser, added --queuelimit 2026-10-10 18:19:37 +08:00
Concedo
aab4e44e89 improve android launch 2026-10-10 17:41:27 +08:00
Xuan-Son Nguyen
781dbc5ac9
spec: properly handle mtmd input for mtp (#30257) 
* spec: properly handle mtmd input for mtp

* nits
 2026-10-10 11:22:38 +02:00
Xuan-Son Nguyen
0fd868cbca
mtmd: add build_inp_attn_mask (#30259) 2026-10-10 11:22:12 +02:00
David Friehs
1623d8ce47
cuda: always use MMVQ for MUL_MAT_ID on sm_60 (#27828) 2026-10-10 09:15:42 +03:00
uvos
1bb2b9fcbe
CUDA/HIP: fix race in flash_attn_ext_f16_process_tile when nbatch_combine != DKQ/2 (#30103) 
Suggested-by: Johannes Gäßler <johannesg@5d6.de>
 2026-10-10 09:14:51 +03:00
Concedo
f7930b00a5 compacting resets usage counters 2026-10-10 14:09:27 +08:00
Aaron Teo
2bbca8f202
ggml-cpu: vectorize fp32 to fp16 conversion (#30157) 
ggml-cpu(s390x): rename ulong to uint64_t sized types



ggml-cpu(s390x): rm comment

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
 2026-10-10 09:08:52 +03:00
Amadeus dyw
404f557b5b
vulkan: use 4 rows for NVIDIA MUL_MAT_ID MMVQ (#29274) 
Keep the existing RDNA3/4 policy unchanged and update only rm_id to use 4 rows for NVIDIA except pre-Turing.
 2026-10-10 09:08:19 +03:00
Aaron Teo
b797c82c7d
ggml: fix s390x all cpu build (#30140) 
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
 2026-10-10 09:07:52 +03:00
Concedo
c496d2d1a8 Merge commit 'a518119d30' into concedo_experimental 
# Conflicts:
#	.github/workflows/release-publish.yml
#	docs/build.md
#	ggml/src/ggml-musa/CMakeLists.txt
#	ggml/src/ggml-webgpu/ggml-webgpu.cpp
#	ggml/src/ggml-webgpu/wgsl-shaders/concat.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/cumsum.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/glu.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/pad.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/repeat.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/rms_norm_mul.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/rope.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/row_norm.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/set.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/set_rows.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/soft_max.wgsl
#	ggml/src/ggml-webgpu/wgsl-shaders/sum_rows.wgsl
 2026-10-10 13:39:58 +08:00
Concedo
b3284ea444 Merge commit 'b42b7e6d30' into concedo_experimental 
# Conflicts:
#	.devops/cann.Dockerfile
#	.devops/cpu.Dockerfile
#	.devops/cuda.Dockerfile
#	.devops/intel.Dockerfile
#	.devops/musa.Dockerfile
#	.devops/openvino.Dockerfile
#	.devops/rocm.Dockerfile
#	.devops/s390x.Dockerfile
#	.devops/vulkan.Dockerfile
#	.devops/zendnn.Dockerfile
#	docs/backend/ZenDNN.md
#	docs/function-calling.md
#	examples/json_schema_pydantic_example.py
#	examples/model-conversion/scripts/causal/modelcard.template
#	examples/model-conversion/scripts/embedding/modelcard.template
#	examples/model-conversion/scripts/utils/curl-embedding-server.sh
#	examples/pydantic_models_to_grammar_examples.py
#	ggml/src/ggml-hexagon/htp/im2col-ops.c
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	ggml/src/ggml-sycl/mmvq.cpp
#	ggml/src/ggml-sycl/mmvq.hpp
#	ggml/src/ggml-sycl/vecdotq.hpp
#	scripts/compare-logprobs.py
#	scripts/hip/gcn-cdna-vgpr-check.py
#	scripts/server-bench.py
#	scripts/server-test-function-call.py
#	scripts/server-test-model.py
#	scripts/server-test-parallel-tc.py
#	scripts/server-test-structured.py
#	tests/test-backend-ops.cpp
#	tests/test-llama-archs.cpp
#	tools/cli/README.md
#	tools/completion/README.md
#	tools/server/README.md
#	tools/server/bench/README.md
 2026-10-10 13:37:53 +08:00