-
483609509d
ui: add opt-in run_javascript frontend tool (#24244)
Pascal
2026-06-09 18:02:31 +02:00
-
49f3542190
mtmd: build_vit batching (#24352)
Saba Fallah
2026-06-09 16:32:08 +02:00
-
d6d0ce8215
vulkan: reduce iq1 shared memory usage for mul_mm (#24287)
Jeff Bolz
2026-06-09 06:27:38 -05:00
-
b4e3dc613b
vulkan: add
v_dot2_f32_f16 support in matrix-matrix multiplication and Flash Attention (#24123)
Ruben Ortlam
2026-06-09 13:27:04 +02:00
-
ae735b1314
ui: Fix excessive style recalculation on hover (#24243)
Nick Towle
2026-06-09 03:52:20 -07:00
-
9682e351b8
mtmd: refactor video subproc handling (#24316)
Xuan-Son NguyenandMikko Juola
2026-06-09 12:15:12 +02:00
-
1e912561dd
server: log prompts to directory (#22031)
jacekpoplawskiandXuan-Son Nguyen
2026-06-09 12:09:07 +02:00
-
efbacf8d21
ui: fix mobile chat form overflow and bust stale bundle cache (#24158)
Pascal
2026-06-09 11:12:58 +02:00
-
26021699bc
ggml : add GGML_OP_COL2IM_1D (#24206)
Pascal
2026-06-09 11:01:37 +02:00
-
961e9a3e46
server : do not clear slots without unified KV cache (#24190)
2026-06-09 09:45:16 +02:00
-
f0152efe40
models : fix plamo2 attention_key/value_length regression (#24317)
Sigbjørn Skjæret
2026-06-09 09:26:44 +02:00
-
fd3271e0b4
ggml-cpu : fix rms_norm_back wrong output under in-place aliasing (#24305)
Yash Raj PandeyandGeorgi Gerganov
2026-06-09 03:24:27 -04:00
-
e3471b3e73
Remove case for GGML_TYPE_Q4_K in mvvq.cu (#23528)
ravel7524
2026-06-09 07:46:23 +02:00
-
3ac3c20c96
ggml-webgpu: Add clang-format job (#24308)
Reese Levine
2026-06-08 20:54:24 -07:00
-
1e1aca09da
ggml-webgpu: Improve prefill speeds for k-quants + refactor matmul for Q4/Q5/Q8 and k-quants (#24225)
Masashi Yoshimura
2026-06-09 07:19:56 +09:00
-
7d2b45b4f7
mtp: support for gemma-4 E2B and E4B assistants (#24282)
Max Krasnyansky
2026-06-08 13:48:52 -07:00
-
42a0afd594
server : do not parse when flushing http headers (#24281)
Aldehir Rojas
2026-06-08 13:32:41 -05:00
-
a66d50588b
graph: guard iswa kq_mask on its own buffer (#24294)
PascalandGeorgi Gerganov
2026-06-08 19:20:28 +02:00
-
1705d434f6
[ggml-webgpu] Handle buffer overlap / buffer aliasing for concat operator (#24000)
2026-06-08 08:07:31 -07:00
-
3b3da01dc2
[ggml-webgpu] Implement 2D workgroups for scale, binary, and unary ops (#24044)
Nikhil Jain
2026-06-08 08:07:15 -07:00
-
3ebe862b5d
docker: install ffmpeg in the released image (#24302)
Xuan-Son Nguyen
2026-06-08 16:59:57 +02:00
-
8f83d6c271
mtmd : add video input support (#24269)
Xuan-Son Nguyen
2026-06-08 13:40:12 +02:00
-
c2b1518fd4
sync : ggml
Georgi Gerganov
2026-06-08 12:56:07 +03:00
-
6a1de6fbf1
ggml : bump version to 0.14.0 (ggml/1533)
Georgi Gerganov
2026-06-08 12:51:59 +03:00
-
715b86a366
cli: fix spinner not show during prompt processing (#24283)
Xuan-Son Nguyen
2026-06-08 11:11:45 +02:00
-
c74759a244
vulkan: Use cm2 decode_vector for mul_mat_id B matrix loads (#23991)
Jeff Bolz
2026-06-08 03:40:37 -05:00
-
0f7fada56b
cuda: reset cuda context after reading memory size (#23935)
Ruben Ortlam
2026-06-08 10:22:44 +02:00
-
19bba67c1f
HIP: add gfx1152 and gfx1153 to RDNA3.5 (#24129)
Harkirat Gill
2026-06-08 02:33:23 -04:00
-
daf6bc9f2d
metal : fix im2col 1D case (audio models) (#24220)
Xuan-Son Nguyen
2026-06-08 08:03:18 +02:00
-
d403f00ec3
[SYCL] Update compute runtime version to 26.x in docker (#24070)
Neo Zhang
2026-06-08 10:35:18 +08:00
-
9e3b928fd8
common : relax sampler name matching (#23744)
ddh0andSigbjørn Skjæret
2026-06-07 15:48:11 -05:00
-
8a963fc10e
convert : fix conversion for Mistral-Medium-3.5-128B (#24268)
David Friehs
2026-06-07 21:41:39 +02:00
-
379ac6673b
kv-cache : avoid kv cells copies (#24277)
Georgi Gerganov
2026-06-07 21:42:54 +03:00
-
f0156d1401
kv-cache: follow the source cache size when sharing cells (#24267)
Pascal
2026-06-07 17:33:00 +02:00
-
04eb4c446d
llama : add Gemma4 MTP (#23398)
Aman Gupta
2026-06-07 20:50:54 +08:00
-
8a091c47ab
spec : fix vocab compatibility check (#24256)
Sigbjørn Skjæret
2026-06-07 13:43:52 +02:00
-
465b1f0e75
arg: Skip mmproj download when user supplied mmproj (#24239)
konradmb
2026-06-07 11:18:44 +02:00
-
f71af352a5
convert : fix Gemma4 with no audio encoder (#24242)
Sigbjørn Skjæret
2026-06-07 08:43:05 +02:00
-
3f7c79d7b5
docker : bump cuda13 to 13.3.0 (#24228)
Sigbjørn Skjæret
2026-06-07 08:31:58 +02:00
-
98d5e8ba8a
common/chat : fix LFM2/LFM2.5 reasoning round-trip and <think> leak (#24234)
Tarek Dakhran
2026-06-06 22:39:21 +02:00
-
31e82494c0
mtmd: support "frame merge" for qwen-vl-based models (#21858)
Xuan-Son Nguyenandandrewmd5
2026-06-06 21:17:25 +02:00
-
6b80c74f28
completion : remove useless statics (#24226)
Adrien Gallouët
2026-06-06 12:16:16 +02:00
-
588f0dc2ce
completion : fix format specifier in LOG_INF (#24213)
Adrien Gallouët
2026-06-06 11:24:27 +02:00
-
f5c6ae1827
mtmd, server: add "placeholder bitmap" for counting tokens , add */input_tokens API (#23913)
Xuan-Son Nguyen
2026-06-06 11:06:51 +02:00
-
5a69c97439
vulkan: check coopmat2 features before reporting support (#24186)
Ruben Ortlam
2026-06-06 09:11:35 +02:00
-
5343f4502a
model : rename local n_layer_all variable (#24209)
Sigbjørn Skjæret
2026-06-06 06:07:20 +02:00
-
603300b008
context : fix off-by-one comparisons to n_gpu_layers (#24208)
Sigbjørn Skjæret
2026-06-06 06:06:47 +02:00
-
308f61c31f
opencl: improve get_rows, cpy, concat and q6_k flat gemv (#24160)
lhez
2026-06-05 13:45:25 -07:00
-
da87e9b612
common/chat : unify and fix LFM2/LFM2.5 tool parser (#24178)
Tarek Dakhran
2026-06-05 21:31:56 +02:00
-
e82beaa60d
vulkan: add fwht support for Intel with shmem reduction (#23964)
Ruben Ortlam
2026-06-05 19:44:40 +02:00
-
c4a278d68e
model: fix build failed (#24193)
Xuan-Son Nguyen
2026-06-05 18:12:27 +02:00
-
64086f2b2f
model, mtmd: Granite4 Vision (#23545)
Gabe Goodhart
2026-06-05 09:44:59 -06:00
-
6effcecd0b
TP: round up granularity to 128 (#24180)
Johannes Gäßler
2026-06-05 17:35:13 +02:00
-
86591c7536
cli: fix model params not propagated (#23893)
therealkenc
2026-06-05 08:29:41 -07:00
-
96fbe00393
model : fix llama_model::n_gpu_layers() (#24188)
Georgi Gerganov
2026-06-05 17:11:42 +03:00
-
2016bf2b3b
ui: run npm install when package-lock.json is newer than node_modules (#24171)
Pascal
2026-06-05 14:57:32 +02:00
-
9c955c48b0
Fix link to available UI settings (#24169)
Mario
2026-06-05 13:39:32 +01:00
-
cc7bef34e2
ui: add ignore-scripts=true to npmrc (#24149)
Xuan-Son Nguyen
2026-06-05 14:31:03 +02:00
-
ad1b88ca0d
docs: Update quantization readme (#24133)
Pedro CuencaandSigbjørn Skjæret
2026-06-05 12:21:26 +02:00
-
59917d3922
minor : fix lint issues (#24165)
Georgi Gerganov
2026-06-05 11:17:54 +03:00
-
7acb4e8cd2
hparams : refactor
hparams.n_layer (#24060)
Georgi GerganovandSigbjørn Skjæret
2026-06-05 11:09:36 +03:00
-
3ecfb150a4
kleidiai : dynamic chunck-based scheduling for hybrid execution (#23819)
Charles Xu
2026-06-05 09:11:47 +02:00
-
2154a0fdcf
CUDA: enroll mul_mat_vec_q_moe into pdl (#24087)
Oliver Simons
2026-06-05 08:37:34 +02:00
-
46fa662b1f
ci : build-msys job slimming [no ci] (#24157)
Daniel Bevenius
2026-06-05 07:57:36 +02:00
-
7fe2ae45ab
sycl : port multi-column MMVQ from CUDA backend (#21845)
Mason Milburn
2026-06-05 01:10:31 -04:00
-
7c158fbb4a
server : disable on-device spec checkpoints (#24108)
Georgi Gerganov
2026-06-04 19:30:59 +03:00
-
260862b8ca
arg: fix double mtp downloads (#24128)
Xuan-Son Nguyen
2026-06-04 18:23:48 +02:00
-
42b2d60e57
webui: [a11y] fix keyboard navigation issues in chat interface and sidebar (#23132)
viggy
2026-06-04 08:59:00 -07:00
-
e7bcf1c3a8
Move duplicated imatrix code into single common imatrix-loader.cpp (#22445)
Bartowski
2026-06-04 11:45:40 -04:00
-
21444c822e
ui: Fixed packages (#24119)
Aleksander Grygier
2026-06-04 16:23:08 +02:00
-
526977068f
ui: added single line reasoning preview (#23601)
MagicExistsandAleksander Grygier
2026-06-04 21:09:43 +07:00
-
0dbfa66a1f
return filter to save memory (#24125)
forforever73andlvyichen
2026-06-04 21:56:33 +08:00
-
e8023568d0
convert: Fix Gemma 4 Unified conversion (#24118)
Pedro Cuenca
2026-06-04 15:21:38 +02:00
-
4c51309617
ggml: vectorize ggml_vec_dot_q4_1_q8_1 with WASM SIMD128 (#22209)
Kartik Sirohi
2026-06-04 18:42:38 +05:30
-
6f3a9f3dee
server: avoid unnecessary checkpoint restore when new tokens are present (#24110)
Yongyue SunandGeorgi Gerganov
2026-06-04 21:09:01 +08:00
-
a121232fdc
agents: refactor, include more guidelines (#24111)
Xuan-Son Nguyen
2026-06-04 13:40:23 +02:00
-
4586479852
webui: fix tool selector toggle/counter, key tools by stable identity (#24065)
Pascal
2026-06-04 13:09:49 +02:00
-
4d742877b2
build : use umbrella Headers directory for XCFramework module map (#23974)
Gerard Martinez
2026-06-04 03:58:25 -07:00
-
0066404085
server : add header to tools/server/server-http.h (#24089)
A B
2026-06-04 05:14:46 -05:00
-
7ac5a4225e
cmake: skip cvector-generator and export-lora when CPU backend is disabled (#24053)
Andrea Richiardi
2026-06-04 04:13:19 -06:00
-
e3ba22d6cc
fix(mtmd): handle Gemma 4 audio projector embedding size (#24091)
AndreiandXuan Son Nguyen
2026-06-04 02:51:23 -07:00
-
6ddc9430b1
readme : add status badges (#24104)
Georgi Gerganov
2026-06-04 10:58:13 +03:00
-
65ef50a0a4
tests : refactor test-save-load-state to accept token input (#24073)
Georgi Gerganov
2026-06-04 08:06:36 +03:00
-
3d1998634e
metal : reduce rset heartbeat from 500ms -> 5ms (#24074)
Georgi Gerganov
2026-06-04 08:05:32 +03:00
-
e8c54893f2
ggml-webgpu: FlashAttention refactor + standardize quantization support (#23834)
Reese Levine
2026-06-03 22:05:04 -07:00
-
3c7450cee1
ggml-cpu: extend RVV quantization vec dot to higher VLENs (#22754)
2026-06-04 10:03:40 +05:00
-
f478f1b6d7
sycl : Improve SYCL doc (#23025)
Todd Malsbary
2026-06-03 22:02:54 -07:00
-
94a220cd67
mtmd: fix Gemma 4 unified FPE (#24088)
Andrei
2026-06-03 12:51:18 -07:00
-
166fe29492
qwen35: use post-norm hidden state for MTP (#24025)
Aman Gupta
2026-06-04 01:29:09 +08:00
-
c8d6a00636
mtmd: enable non-causal vision for gemma 4 unified (#24082)
Xuan-Son Nguyen
2026-06-03 19:05:17 +02:00
-
a731805ced
mtmd, model: allow skip build_vit() (#24077)
Xuan-Son Nguyen
2026-06-03 17:10:35 +02:00
-
ee4cf705bb
ui: Mermaid Diagrams in chat + interactive preview (#24032)
Aleksander Grygier
2026-06-03 16:55:36 +02:00
-
9e58d4d692
Avoid PDL race conditions by disabling __restrict__ when PDL is used (#24030)
Andreas KieslingerandOliver Simons
2026-06-03 13:56:42 +02:00
-
3571fa5435
ggml-cpu: use runtime SVE width in FWHT (#24059)
Charles Xu
2026-06-03 12:45:10 +02:00
-
f8f0a47a55
cuda: reserve space for quantize kv-cache at startup (#23907)
Aman GuptaandJohannes Gäßler
2026-06-03 18:39:59 +08:00
-
06938ac129
tests : add support for qwen3 SSM archs (#24031)
Georgi Gerganov
2026-06-03 10:15:27 +03:00
-
d545a2a993
update BoringSSL to 0.20260526.0 (#23794)
Alessandro de Oliveira Faria (A.K.A.CABELO)
2026-06-03 02:42:58 -03:00
-
4da6370d43
ci : disable ccache for msvc windows release jobs (#23911)
Georgi Gerganov
2026-06-03 08:05:21 +03:00
-
e3666269f9
arg : removed unecesary mmproj download when users pass --no-mmproj (#23425)
Ryan Mangeno
2026-06-02 22:04:46 -07:00
-
63e66fdd23
opencl: use flat variants of q4_K and q6_K gemv for very large M (#24006)
lhez
2026-06-02 14:16:17 -07:00