Georgi Gerganov
|
d2fcd91cf9
server : disable context shift by default (#15416)
|
5 месяцев назад |
SHUAI YANG
|
a6d3cfe7fa
CANN: optimize rope operator (#15335)
|
5 месяцев назад |
R0CKSTAR
|
67f09a3a27
musa: handle __hgt2_mask, available starting from MUSA SDK rc4.3.0 (#15413)
|
5 месяцев назад |
Marvin Gießing
|
6424594c56
ggml-cpu: add mxfp4 VSX intrinsics for Power9+ (ppc64le) hardware (#15385)
|
5 месяцев назад |
Xuan-Son Nguyen
|
e9288e8869
chat : clarify the meaning of reasoning_format (#15408)
|
5 месяцев назад |
Georgi Gerganov
|
9d262f4bad
server : remove swa_full warning (#15399)
|
5 месяцев назад |
Georgi Gerganov
|
f0d3c7405c
batched-bench : use rand tokens (#15398)
|
5 месяцев назад |
Xuan-Son Nguyen
|
f08c4c0d8d
mtmd : clean up clip_n_output_tokens (#15391)
|
5 месяцев назад |
Georgi Gerganov
|
6d7f1117e3
codeowners : remove mmv.*
|
5 месяцев назад |
Georgi Gerganov
|
60212f1ead
sync : ggml
|
5 месяцев назад |
Georgi Gerganov
|
f0c541d315
scripts : update sync scripts
|
5 месяцев назад |
Sigbjørn Skjæret
|
baa9255a45
llama : merge conts and reshapes and remove unnecessary cont (#15380)
|
5 месяцев назад |
Georgi Gerganov
|
3007baf201
readme : update hot topics (#15397)
|
5 месяцев назад |
davidef
|
d1d8241600
server : fix incoming tasks not process in order (#15395)
|
5 месяцев назад |
Dobri Danchev
|
618575c582
Fix broken build: require updated pip to support --break-system-packages (#15357)
|
5 месяцев назад |
compilade
|
f44f793172
ggml-quants : fix make_qp_quants NANs and IQ1 assertion errors (#15379)
|
5 месяцев назад |
Jeff Bolz
|
ae532eac2c
vulkan: disable spirv-opt for bfloat16 shaders (#15352)
|
5 месяцев назад |
Oleksandr Kuvshynov
|
e5155e6986
server : export max observed n_past value (#15361)
|
5 месяцев назад |
Jeff Bolz
|
21c17b5bef
vulkan: Use larger workgroups for mul_mat_vec when M is small (#15355)
|
5 месяцев назад |
Dong Won Kim
|
19f4decae0
vulkan: support sqrt (#15370)
|
5 месяцев назад |
Sigbjørn Skjæret
|
4d196981d4
convert : force patch_embd weights to F16 or F32 to avoid broken GGUFs (#15367)
|
5 месяцев назад |
Sigbjørn Skjæret
|
b143fbc87a
ci : fix hang in windows-hip build/release (#15365)
|
5 месяцев назад |
Jeff Bolz
|
de5627910d
vulkan: Optimize argsort (#15354)
|
5 месяцев назад |
Tarek Dakhran
|
65349f26f2
model : support vision LiquidAI LFM2-VL family (#15347)
|
5 месяцев назад |
Jeff Bolz
|
1fe00296f5
vulkan: fuse adds (#15252)
|
5 месяцев назад |
Jeff Bolz
|
de2192794f
vulkan: Support mul_mat_id with f32 accumulators (#15337)
|
5 месяцев назад |
Jeff Bolz
|
2e2b22ba66
vulkan: Add missing bounds checking to scalar/coopmat1 mul_mat_id (#15334)
|
5 месяцев назад |
rmatif
|
912ff8c119
OpenCL: add initial FA support (#14987)
|
5 месяцев назад |
Daniel Bevenius
|
5e6229a840
common : fix double bos, use common_chat_templates for add_bos and add_eos (#15326)
|
5 месяцев назад |
lhez
|
e2c1bfff53
opencl: add initial mxfp4 support via mv (#15270)
|
5 месяцев назад |