| Name | Mode | Size | Raw | |
|---|---|---|---|---|
| .. | ||||
| d | template-instances/ | 0040000 |
||
| d | vendors/ | 0040000 |
||
| f | CMakeLists.txt | 0100644 |
10.6 KB |
raw |
| f | acc.cu | 0100644 |
2.4 KB |
raw |
| f | acc.cuh | 0100644 |
131 B |
raw |
| f | add-id.cu | 0100644 |
1.9 KB |
raw |
| f | add-id.cuh | 0100644 |
101 B |
raw |
| f | arange.cu | 0100644 |
1.2 KB |
raw |
| f | arange.cuh | 0100644 |
137 B |
raw |
| f | argmax.cu | 0100644 |
2.8 KB |
raw |
| f | argmax.cuh | 0100644 |
98 B |
raw |
| f | argsort.cu | 0100644 |
9.2 KB |
raw |
| f | argsort.cuh | 0100644 |
868 B |
raw |
| f | binbcast.cu | 0100644 |
21.4 KB |
raw |
| f | binbcast.cuh | 0100644 |
579 B |
raw |
| f | clamp.cu | 0100644 |
1.6 KB |
raw |
| f | clamp.cuh | 0100644 |
135 B |
raw |
| f | common.cuh | 0100644 |
51.4 KB |
raw |
| f | concat.cu | 0100644 |
7.4 KB |
raw |
| f | concat.cuh | 0100644 |
137 B |
raw |
| f | conv-transpose-1d.cu | 0100644 |
3.2 KB |
raw |
| f | conv-transpose-1d.cuh | 0100644 |
158 B |
raw |
| f | conv2d-dw.cu | 0100644 |
7.2 KB |
raw |
| f | conv2d-dw.cuh | 0100644 |
155 B |
raw |
| f | conv2d-transpose.cu | 0100644 |
3.7 KB |
raw |
| f | conv2d-transpose.cuh | 0100644 |
157 B |
raw |
| f | conv2d.cu | 0100644 |
6.4 KB |
raw |
| f | conv2d.cuh | 0100644 |
149 B |
raw |
| f | convert.cu | 0100644 |
30.6 KB |
raw |
| f | convert.cuh | 0100644 |
1.9 KB |
raw |
| f | count-equal.cu | 0100644 |
2.0 KB |
raw |
| f | count-equal.cuh | 0100644 |
144 B |
raw |
| f | cp-async.cuh | 0100644 |
2.1 KB |
raw |
| f | cpy-utils.cuh | 0100644 |
6.2 KB |
raw |
| f | cpy.cu | 0100644 |
27.5 KB |
raw |
| f | cpy.cuh | 0100644 |
227 B |
raw |
| f | cross-entropy-loss.cu | 0100644 |
5.9 KB |
raw |
| f | cross-entropy-loss.cuh | 0100644 |
251 B |
raw |
| f | cumsum.cu | 0100644 |
10.7 KB |
raw |
| f | cumsum.cuh | 0100644 |
137 B |
raw |
| f | dequantize.cuh | 0100644 |
2.0 KB |
raw |
| f | diag.cu | 0100644 |
2.6 KB |
raw |
| f | diag.cuh | 0100644 |
133 B |
raw |
| f | diagmask.cu | 0100644 |
1.7 KB |
raw |
| f | diagmask.cuh | 0100644 |
150 B |
raw |
| f | fattn-common.cuh | 0100644 |
37.1 KB |
raw |
| f | fattn-mma-f16.cuh | 0100644 |
81.5 KB |
raw |
| f | fattn-tile.cu | 0100644 |
1.7 KB |
raw |
| f | fattn-tile.cuh | 0100644 |
56.6 KB |
raw |
| f | fattn-vec.cuh | 0100644 |
23.1 KB |
raw |
| f | fattn-wmma-f16.cu | 0100644 |
27.5 KB |
raw |
| f | fattn-wmma-f16.cuh | 0100644 |
2.0 KB |
raw |
| f | fattn.cu | 0100644 |
18.6 KB |
raw |
| f | fattn.cuh | 0100644 |
185 B |
raw |
| f | fill.cu | 0100644 |
1.1 KB |
raw |
| f | fill.cuh | 0100644 |
99 B |
raw |
| f | getrows.cu | 0100644 |
12.5 KB |
raw |
| f | getrows.cuh | 0100644 |
622 B |
raw |
| f | ggml-cuda.cu | 0100644 |
205.0 KB |
raw |
| f | gla.cu | 0100644 |
3.0 KB |
raw |
| f | gla.cuh | 0100644 |
112 B |
raw |
| f | im2col.cu | 0100644 |
12.2 KB |
raw |
| f | im2col.cuh | 0100644 |
218 B |
raw |
| f | mean.cu | 0100644 |
2.8 KB |
raw |
| f | mean.cuh | 0100644 |
99 B |
raw |
| f | mma.cuh | 0100644 |
56.3 KB |
raw |
| f | mmf.cu | 0100644 |
7.7 KB |
raw |
| f | mmf.cuh | 0100644 |
44.0 KB |
raw |
| f | mmid.cu | 0100644 |
7.8 KB |
raw |
| f | mmid.cuh | 0100644 |
258 B |
raw |
| f | mmq.cu | 0100644 |
14.1 KB |
raw |
| f | mmq.cuh | 0100644 |
168.0 KB |
raw |
| f | mmvf.cu | 0100644 |
39.1 KB |
raw |
| f | mmvf.cuh | 0100644 |
826 B |
raw |
| f | mmvq.cu | 0100644 |
38.4 KB |
raw |
| f | mmvq.cuh | 0100644 |
701 B |
raw |
| f | norm.cu | 0100644 |
27.7 KB |
raw |
| f | norm.cuh | 0100644 |
838 B |
raw |
| f | opt-step-adamw.cu | 0100644 |
2.9 KB |
raw |
| f | opt-step-adamw.cuh | 0100644 |
150 B |
raw |
| f | opt-step-sgd.cu | 0100644 |
1.7 KB |
raw |
| f | opt-step-sgd.cuh | 0100644 |
146 B |
raw |
| f | out-prod.cu | 0100644 |
2.3 KB |
raw |
| f | out-prod.cuh | 0100644 |
100 B |
raw |
| f | pad.cu | 0100644 |
4.4 KB |
raw |
| f | pad.cuh | 0100644 |
131 B |
raw |
| f | pad_reflect_1d.cu | 0100644 |
3.4 KB |
raw |
| f | pad_reflect_1d.cuh | 0100644 |
153 B |
raw |
| f | pool2d.cu | 0100644 |
3.2 KB |
raw |
| f | pool2d.cuh | 0100644 |
137 B |
raw |
| f | quantize.cu | 0100644 |
12.4 KB |
raw |
| f | quantize.cuh | 0100644 |
1.7 KB |
raw |
| f | reduce_rows.cuh | 0100644 |
1.1 KB |
raw |
| f | roll.cu | 0100644 |
2.4 KB |
raw |
| f | roll.cuh | 0100644 |
133 B |
raw |
| f | rope.cu | 0100644 |
30.8 KB |
raw |
| f | rope.cuh | 0100644 |
322 B |
raw |
| f | scale.cu | 0100644 |
1.4 KB |
raw |
| f | scale.cuh | 0100644 |
135 B |
raw |
| f | set-rows.cu | 0100644 |
12.9 KB |
raw |
| f | set-rows.cuh | 0100644 |
155 B |
raw |
| f | set.cu | 0100644 |
1.2 KB |
raw |
| f | set.cuh | 0100644 |
145 B |
raw |
| f | softcap.cu | 0100644 |
1.3 KB |
raw |
| f | softcap.cuh | 0100644 |
158 B |
raw |
| f | softmax.cu | 0100644 |
16.9 KB |
raw |
| f | softmax.cuh | 0100644 |
228 B |
raw |
| f | solve_tri.cu | 0100644 |
11.3 KB |
raw |
| f | solve_tri.cuh | 0100644 |
104 B |
raw |
| f | ssm-conv.cu | 0100644 |
6.3 KB |
raw |
| f | ssm-conv.cuh | 0100644 |
103 B |
raw |
| f | ssm-scan.cu | 0100644 |
16.2 KB |
raw |
| f | ssm-scan.cuh | 0100644 |
103 B |
raw |
| f | sum.cu | 0100644 |
1.3 KB |
raw |
| f | sum.cuh | 0100644 |
210 B |
raw |
| f | sumrows.cu | 0100644 |
1.8 KB |
raw |
| f | sumrows.cuh | 0100644 |
212 B |
raw |
| f | top-k.cu | 0100644 |
4.0 KB |
raw |
| f | top-k.cuh | 0100644 |
100 B |
raw |
| f | topk-moe.cu | 0100644 |
15.7 KB |
raw |
| f | topk-moe.cuh | 0100644 |
1013 B |
raw |
| f | tri.cu | 0100644 |
5.2 KB |
raw |
| f | tri.cuh | 0100644 |
131 B |
raw |
| f | tsembd.cu | 0100644 |
1.8 KB |
raw |
| f | tsembd.cuh | 0100644 |
161 B |
raw |
| f | unary.cu | 0100644 |
18.4 KB |
raw |
| f | unary.cuh | 0100644 |
4.0 KB |
raw |
| f | upscale.cu | 0100644 |
13.1 KB |
raw |
| f | upscale.cuh | 0100644 |
139 B |
raw |
| f | vecdotq.cuh | 0100644 |
42.1 KB |
raw |
| f | wkv.cu | 0100644 |
6.9 KB |
raw |
| f | wkv.cuh | 0100644 |
218 B |
raw |