 Douglas Hanley
|
2891c8aa9a
Add support for BERT embedding models (#5423)
|
2 lat temu |
 github-actions[bot]
|
97a336507e
flake.lock: Update
|
2 lat temu |
 Sergio López
|
c88c74f967
vulkan: only use M-sized matmul on Apple GPUs (#5412)
|
2 lat temu |
 Alexey Parfenov
|
a803333a4e
common : use enums for sampler types (#5418)
|
2 lat temu |
|
Alexey Parfenov
|
684780141a
server : allow to specify tokens as strings in logit_bias (#5003)
|
2 lat temu |
 Georgi Gerganov
|
85910c5b30
main : ctrl+C print timing in non-interactive mode (#3873)
|
2 lat temu |
|
Georgi Gerganov
|
139b62a839
common : fix compile warning
|
2 lat temu |
|
Georgi Gerganov
|
0f2411f154
ggml : fix compile warnings (unused vars) (#4966)
|
2 lat temu |
 snadampal
|
a07d0fee1f
ggml : add mmla kernels for quantized GEMM (#4966)
|
2 lat temu |
 Johannes Gäßler
|
e4640d8fdf
lookup: add print for drafting performance (#5450)
|
2 lat temu |
 Xuan Son Nguyen
|
907e08c110
server : add llama2 chat template (#5425)
|
2 lat temu |
 Ian Bull
|
f026f8120f
metal : use autoreleasepool to avoid memory leaks (#5437)
|
2 lat temu |
|
Georgi Gerganov
|
cd9aea63b5
scripts : update sync scripts with new backends
|
2 lat temu |
|
Georgi Gerganov
|
43b65f5eb8
sync : ggml
|
2 lat temu |
 Michael Podvitskiy
|
4633d93af0
ggml : add abort_callback for cpu backend (ggml/725)
|
2 lat temu |
 Neuman Vong
|
4b7b38bef5
vulkan: Set limit for task concurrency (#5427)
|
2 lat temu |
 Daniel Bevenius
|
e00d2a62dd
llava : add requirements.txt and update README.md (#5428)
|
2 lat temu |
 Riley Stewart
|
7c777fcd5d
server : fix prompt caching for repeated prompts (#5420)
|
2 lat temu |
 Paul Tsochantaris
|
e5ca3937c6
llama : do not cap thread count when MoE on CPU (#5419)
|
2 lat temu |
 Marko Tasic
|
e4124c2477
readme : add JavaScript/Wasm repo (#5415)
|
2 lat temu |
|
Michael Podvitskiy
|
b2f87cb64d
ggml : fix `error C2078: too many initializers` for MSVC ARM64 (#5404)
|
2 lat temu |
 0cc4m
|
44fbe34360
Fix Vulkan crash on APUs with very little device memory (#5424)
|
2 lat temu |
|
Johannes Gäßler
|
8e6a9d2de0
CUDA: more warps for mmvq on NVIDIA (#5394)
|
2 lat temu |
 slaren
|
41f308f58e
llama : do not print "offloading layers" message in CPU-only builds (#5416)
|
2 lat temu |
 Abhilash Majumder
|
6e99f2a04f
Fix f16_sycl cpy call from Arc (#5411)
|
2 lat temu |
|
Daniel Bevenius
|
ff4ff05c5f
llava : add missing .py, and fix paths in README.md (#5414)
|
2 lat temu |
|
Johannes Gäßler
|
b7b74cef36
fix trailing whitespace (#5407)
|
2 lat temu |
 runfuture
|
4aa43fab56
llama : fix MiniCPM (#5392)
|
2 lat temu |
|
Daniel Bevenius
|
a6e514a85f
llava: fix typo/formatting in README.md (#5405)
|
2 lat temu |
|
Johannes Gäßler
|
26d4efd11e
sampling: fix top_k <= 0 (#5388)
|
2 lat temu |
