 lhez
|
52e5d421f1
opencl: fix rms_norm_mul (#17250)
|
hai 2 meses |
 shaofeiqi
|
4db5641210
opencl: add kernel to handle mat mul in attention to improve encoding speed (#17181)
|
hai 2 meses |
 shani-f
|
72bd7321a7
sycl : unify unary kernels with a generic implementation and enable wide operator support (#17213)
|
hai 2 meses |
 Aleksander Grygier
|
22e1ce2f81
webui: Fix clickability around chat processing statistics UI (#17278)
|
hai 2 meses |
 Pascal
|
1411d9275a
webui: add OAI-Compat Harmony tool-call streaming visualization and persistence in chat UI (#16618)
|
hai 2 meses |
 Sigbjørn Skjæret
|
662192e1dc
convert : remove unnecessary chat template patching (#17289)
|
hai 2 meses |
 Jeff Bolz
|
24dc769f1b
vulkan: Fuse mul_mat_id+add_id+mul and mul_mat+add+add. (#17287)
|
hai 2 meses |
 Ruben Ortlam
|
4dca015b7e
vulkan: Replace 16-bit unpack8 calls to work around legacy Windows AMD driver bug (#17285)
|
hai 2 meses |
|
Sigbjørn Skjæret
|
9a8860cf5d
convert : use all parts in safetensors index (#17286)
|
hai 2 meses |
|
Sigbjørn Skjæret
|
9d3ef4809f
convert : set expert gating func in base class (#17279)
|
hai 2 meses |
 Ankur Verma
|
c7b7db0445
mtmd-cli: Avoid logging to stdout for model loading messages in mtmd-cli (#17277)
|
hai 2 meses |
 Giuseppe Scrivano
|
1568d13c2c
vulkan: implement ABS and NEG (#17245)
|
hai 2 meses |
|
Jeff Bolz
|
439342ea0b
vulkan: Use ggml_vk_tensor_subbuffer in mul_mat_vec(id) paths (#17244)
|
hai 2 meses |
|
Jeff Bolz
|
234ae7d7bd
vulkan: skip all-negative-inf blocks in FA (#17186)
|
hai 2 meses |
|
Jeff Bolz
|
38eaf32af1
vulkan: change graph_compute to be async and enable get_tensor_async (#17158)
|
hai 2 meses |
 Xuan-Son Nguyen
|
9b17d74ab7
mtmd: add mtmd_log_set (#17268)
|
hai 2 meses |
 Bartowski
|
e1fcf8b09b
model : add AfmoeForCausalLM support (#16477)
|
hai 2 meses |
 Marek Hradil jr.
|
6cd0cf72ce
fix : Dangling pointer for non-empty trigger words in lazy grammar construction (#17048)
|
hai 2 meses |
 Georgi Gerganov
|
d396b43748
server : fix "can batch with" bug (#17263)
|
hai 2 meses |
|
Georgi Gerganov
|
45c6ef7307
metal : support argsort for ne00 > 1024 (#17247)
|
hai 2 meses |
|
Georgi Gerganov
|
2606b0adab
metal : make the FA extra sizes consistent (#17143)
|
hai 2 meses |
 ixgbe
|
307772fcda
readme : add RVV,ZVFH,ZFH,ZICBOP support for RISC-V (#17259)
|
hai 2 meses |
|
Aleksander Grygier
|
f1bad23f88
Better UX for handling multiple attachments in WebUI (#17246)
|
hai 2 meses |
 Alberto Cabrera Pérez
|
becc4816dd
ggml-cpu: handle 3d tensors in repack mat_mul (#17241)
|
hai 2 meses |
|
Xuan-Son Nguyen
|
c4abcb2457
server: fixing naming conflict res_error (#17243)
|
hai 2 meses |
 Piotr Wilkin (ilintar)
|
389ac78b26
ggml : add ops SOFTPLUS, EXPM1, TRI, SOLVE_TRI, CUMSUM (#17063)
|
hai 2 meses |
|
Ruben Ortlam
|
a19bd6f7ce
vulkan: remove shell call from vulkan-shaders-gen tool, revert file check (#17219)
|
hai 2 meses |
 Diego Devesa
|
dd091e52f8
sched : fix reserve ignoring user tensor assignments (#17232)
|
hai 2 meses |
|
ixgbe
|
1215dde7b0
ggml-cpu : add RISC-V vector intrinsic support for silu and cvar operations (#17227)
|
hai 2 meses |
 bagheera
|
0cfb19166b
metal: accelerated conv2d (#17175)
|
hai 2 meses |
