gpustack-runner 0.1.22.post1__py3-none-any.whl → 0.1.22.post3__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- gpustack_runner/_version.py +2 -2
- gpustack_runner/_version_appendix.py +1 -1
- gpustack_runner/runner.py.json +88 -44
- {gpustack_runner-0.1.22.post1.dist-info → gpustack_runner-0.1.22.post3.dist-info}/METADATA +33 -26
- {gpustack_runner-0.1.22.post1.dist-info → gpustack_runner-0.1.22.post3.dist-info}/RECORD +8 -8
- {gpustack_runner-0.1.22.post1.dist-info → gpustack_runner-0.1.22.post3.dist-info}/WHEEL +0 -0
- {gpustack_runner-0.1.22.post1.dist-info → gpustack_runner-0.1.22.post3.dist-info}/entry_points.txt +0 -0
- {gpustack_runner-0.1.22.post1.dist-info → gpustack_runner-0.1.22.post3.dist-info}/licenses/LICENSE +0 -0
gpustack_runner/_version.py
CHANGED
|
@@ -27,8 +27,8 @@ version_tuple: VERSION_TUPLE
|
|
|
27
27
|
__commit_id__: COMMIT_ID
|
|
28
28
|
commit_id: COMMIT_ID
|
|
29
29
|
|
|
30
|
-
__version__ = version = '0.1.22.
|
|
31
|
-
__version_tuple__ = version_tuple = (0, 1, 22, '
|
|
30
|
+
__version__ = version = '0.1.22.post3'
|
|
31
|
+
__version_tuple__ = version_tuple = (0, 1, 22, 'post3')
|
|
32
32
|
try:
|
|
33
33
|
from ._version_appendix import git_commit
|
|
34
34
|
__commit_id__ = commit_id = git_commit
|
|
@@ -1 +1 @@
|
|
|
1
|
-
git_commit = "
|
|
1
|
+
git_commit = "c9f91c3"
|
gpustack_runner/runner.py.json
CHANGED
|
@@ -142,28 +142,6 @@
|
|
|
142
142
|
"docker_image": "gpustack/runner:cann8.2-a3-sglang0.5.1.post3",
|
|
143
143
|
"deprecated": false
|
|
144
144
|
},
|
|
145
|
-
{
|
|
146
|
-
"backend": "cann",
|
|
147
|
-
"backend_version": "8.2",
|
|
148
|
-
"original_backend_version": "8.2.rc2",
|
|
149
|
-
"backend_variant": "a3",
|
|
150
|
-
"service": "vllm",
|
|
151
|
-
"service_version": "0.11.0",
|
|
152
|
-
"platform": "linux/amd64",
|
|
153
|
-
"docker_image": "gpustack/runner:cann8.2-a3-vllm0.11.0",
|
|
154
|
-
"deprecated": false
|
|
155
|
-
},
|
|
156
|
-
{
|
|
157
|
-
"backend": "cann",
|
|
158
|
-
"backend_version": "8.2",
|
|
159
|
-
"original_backend_version": "8.2.rc2",
|
|
160
|
-
"backend_variant": "a3",
|
|
161
|
-
"service": "vllm",
|
|
162
|
-
"service_version": "0.11.0",
|
|
163
|
-
"platform": "linux/arm64",
|
|
164
|
-
"docker_image": "gpustack/runner:cann8.2-a3-vllm0.11.0",
|
|
165
|
-
"deprecated": false
|
|
166
|
-
},
|
|
167
145
|
{
|
|
168
146
|
"backend": "cann",
|
|
169
147
|
"backend_version": "8.2",
|
|
@@ -373,28 +351,6 @@
|
|
|
373
351
|
"docker_image": "gpustack/runner:cann8.2-910b-sglang0.5.1.post3",
|
|
374
352
|
"deprecated": false
|
|
375
353
|
},
|
|
376
|
-
{
|
|
377
|
-
"backend": "cann",
|
|
378
|
-
"backend_version": "8.2",
|
|
379
|
-
"original_backend_version": "8.2.rc2",
|
|
380
|
-
"backend_variant": "910b",
|
|
381
|
-
"service": "vllm",
|
|
382
|
-
"service_version": "0.11.0",
|
|
383
|
-
"platform": "linux/amd64",
|
|
384
|
-
"docker_image": "gpustack/runner:cann8.2-910b-vllm0.11.0",
|
|
385
|
-
"deprecated": false
|
|
386
|
-
},
|
|
387
|
-
{
|
|
388
|
-
"backend": "cann",
|
|
389
|
-
"backend_version": "8.2",
|
|
390
|
-
"original_backend_version": "8.2.rc2",
|
|
391
|
-
"backend_variant": "910b",
|
|
392
|
-
"service": "vllm",
|
|
393
|
-
"service_version": "0.11.0",
|
|
394
|
-
"platform": "linux/arm64",
|
|
395
|
-
"docker_image": "gpustack/runner:cann8.2-910b-vllm0.11.0",
|
|
396
|
-
"deprecated": false
|
|
397
|
-
},
|
|
398
354
|
{
|
|
399
355
|
"backend": "cann",
|
|
400
356
|
"backend_version": "8.2",
|
|
@@ -648,6 +604,28 @@
|
|
|
648
604
|
"docker_image": "gpustack/runner:cuda12.9-sglang0.5.6.post2",
|
|
649
605
|
"deprecated": false
|
|
650
606
|
},
|
|
607
|
+
{
|
|
608
|
+
"backend": "cuda",
|
|
609
|
+
"backend_version": "12.9",
|
|
610
|
+
"original_backend_version": "12.9.1",
|
|
611
|
+
"backend_variant": "",
|
|
612
|
+
"service": "vllm",
|
|
613
|
+
"service_version": "0.13.0",
|
|
614
|
+
"platform": "linux/amd64",
|
|
615
|
+
"docker_image": "gpustack/runner:cuda12.9-vllm0.13.0",
|
|
616
|
+
"deprecated": false
|
|
617
|
+
},
|
|
618
|
+
{
|
|
619
|
+
"backend": "cuda",
|
|
620
|
+
"backend_version": "12.9",
|
|
621
|
+
"original_backend_version": "12.9.1",
|
|
622
|
+
"backend_variant": "",
|
|
623
|
+
"service": "vllm",
|
|
624
|
+
"service_version": "0.13.0",
|
|
625
|
+
"platform": "linux/arm64",
|
|
626
|
+
"docker_image": "gpustack/runner:cuda12.9-vllm0.13.0",
|
|
627
|
+
"deprecated": false
|
|
628
|
+
},
|
|
651
629
|
{
|
|
652
630
|
"backend": "cuda",
|
|
653
631
|
"backend_version": "12.9",
|
|
@@ -769,6 +747,28 @@
|
|
|
769
747
|
"docker_image": "gpustack/runner:cuda12.8-sglang0.5.4.post3",
|
|
770
748
|
"deprecated": false
|
|
771
749
|
},
|
|
750
|
+
{
|
|
751
|
+
"backend": "cuda",
|
|
752
|
+
"backend_version": "12.8",
|
|
753
|
+
"original_backend_version": "12.8.1",
|
|
754
|
+
"backend_variant": "",
|
|
755
|
+
"service": "vllm",
|
|
756
|
+
"service_version": "0.13.0",
|
|
757
|
+
"platform": "linux/amd64",
|
|
758
|
+
"docker_image": "gpustack/runner:cuda12.8-vllm0.13.0",
|
|
759
|
+
"deprecated": false
|
|
760
|
+
},
|
|
761
|
+
{
|
|
762
|
+
"backend": "cuda",
|
|
763
|
+
"backend_version": "12.8",
|
|
764
|
+
"original_backend_version": "12.8.1",
|
|
765
|
+
"backend_variant": "",
|
|
766
|
+
"service": "vllm",
|
|
767
|
+
"service_version": "0.13.0",
|
|
768
|
+
"platform": "linux/arm64",
|
|
769
|
+
"docker_image": "gpustack/runner:cuda12.8-vllm0.13.0",
|
|
770
|
+
"deprecated": false
|
|
771
|
+
},
|
|
772
772
|
{
|
|
773
773
|
"backend": "cuda",
|
|
774
774
|
"backend_version": "12.8",
|
|
@@ -923,6 +923,28 @@
|
|
|
923
923
|
"docker_image": "gpustack/runner:cuda12.8-voxbox0.0.20",
|
|
924
924
|
"deprecated": false
|
|
925
925
|
},
|
|
926
|
+
{
|
|
927
|
+
"backend": "cuda",
|
|
928
|
+
"backend_version": "12.6",
|
|
929
|
+
"original_backend_version": "12.6.3",
|
|
930
|
+
"backend_variant": "",
|
|
931
|
+
"service": "vllm",
|
|
932
|
+
"service_version": "0.13.0",
|
|
933
|
+
"platform": "linux/amd64",
|
|
934
|
+
"docker_image": "gpustack/runner:cuda12.6-vllm0.13.0",
|
|
935
|
+
"deprecated": false
|
|
936
|
+
},
|
|
937
|
+
{
|
|
938
|
+
"backend": "cuda",
|
|
939
|
+
"backend_version": "12.6",
|
|
940
|
+
"original_backend_version": "12.6.3",
|
|
941
|
+
"backend_variant": "",
|
|
942
|
+
"service": "vllm",
|
|
943
|
+
"service_version": "0.13.0",
|
|
944
|
+
"platform": "linux/arm64",
|
|
945
|
+
"docker_image": "gpustack/runner:cuda12.6-vllm0.13.0",
|
|
946
|
+
"deprecated": false
|
|
947
|
+
},
|
|
926
948
|
{
|
|
927
949
|
"backend": "cuda",
|
|
928
950
|
"backend_version": "12.6",
|
|
@@ -1242,6 +1264,17 @@
|
|
|
1242
1264
|
"docker_image": "gpustack/runner:rocm7.0-sglang0.5.6.post2",
|
|
1243
1265
|
"deprecated": false
|
|
1244
1266
|
},
|
|
1267
|
+
{
|
|
1268
|
+
"backend": "rocm",
|
|
1269
|
+
"backend_version": "7.0",
|
|
1270
|
+
"original_backend_version": "7.0.2",
|
|
1271
|
+
"backend_variant": "",
|
|
1272
|
+
"service": "vllm",
|
|
1273
|
+
"service_version": "0.13.0",
|
|
1274
|
+
"platform": "linux/amd64",
|
|
1275
|
+
"docker_image": "gpustack/runner:rocm7.0-vllm0.13.0",
|
|
1276
|
+
"deprecated": false
|
|
1277
|
+
},
|
|
1245
1278
|
{
|
|
1246
1279
|
"backend": "rocm",
|
|
1247
1280
|
"backend_version": "7.0",
|
|
@@ -1297,6 +1330,17 @@
|
|
|
1297
1330
|
"docker_image": "gpustack/runner:rocm6.4-sglang0.5.5.post3",
|
|
1298
1331
|
"deprecated": false
|
|
1299
1332
|
},
|
|
1333
|
+
{
|
|
1334
|
+
"backend": "rocm",
|
|
1335
|
+
"backend_version": "6.4",
|
|
1336
|
+
"original_backend_version": "6.4.4",
|
|
1337
|
+
"backend_variant": "",
|
|
1338
|
+
"service": "vllm",
|
|
1339
|
+
"service_version": "0.13.0",
|
|
1340
|
+
"platform": "linux/amd64",
|
|
1341
|
+
"docker_image": "gpustack/runner:rocm6.4-vllm0.13.0",
|
|
1342
|
+
"deprecated": false
|
|
1343
|
+
},
|
|
1300
1344
|
{
|
|
1301
1345
|
"backend": "rocm",
|
|
1302
1346
|
"backend_version": "6.4",
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: gpustack-runner
|
|
3
|
-
Version: 0.1.22.
|
|
3
|
+
Version: 0.1.22.post3
|
|
4
4
|
Summary: GPUStack Runner is library for registering runnable accelerated backends and services in GPUStack.
|
|
5
5
|
Project-URL: Homepage, https://github.com/gpustack/runner
|
|
6
6
|
Project-URL: Bug Tracker, https://github.com/gpustack/gpustack/issues
|
|
@@ -53,17 +53,18 @@ The following table lists the supported accelerated backends and their correspon
|
|
|
53
53
|
and [#2795](https://github.com/vllm-project/vllm-ascend/issues/2795).
|
|
54
54
|
|
|
55
55
|
> [!IMPORTANT]
|
|
56
|
-
> -
|
|
57
|
-
> -
|
|
58
|
-
|
|
59
|
-
|
|
60
|
-
|
|
61
|
-
|
|
62
|
-
| 8.3 (
|
|
63
|
-
| 8.3 (
|
|
64
|
-
| 8.
|
|
65
|
-
| 8.2 (
|
|
66
|
-
| 8.2 (
|
|
56
|
+
> - Applied [ATB model patched](https://github.com/gpustack/gpustack/issues/2016#issuecomment-3646603380) to MindIE 2.2.rc1/2.1.rc2.
|
|
57
|
+
> - Applied [av package](https://github.com/gpustack/gpustack/issues/2016#issuecomment-3631228085) to MindIE 2.2.rc1/2.1.rc2.
|
|
58
|
+
> - Update vLLM 0.11.0 with stable vLLM Ascend plugin.
|
|
59
|
+
|
|
60
|
+
| CANN Version <br/> (Variant) | MindIE | vLLM | SGLang |
|
|
61
|
+
|------------------------------|--------------------------|--------------------------------------------------------------------------|------------------------|
|
|
62
|
+
| 8.3 (A3/910C) | **`2.2.rc1`** | `0.12.0`, **`0.11.0`** | `0.5.6.post2` |
|
|
63
|
+
| 8.3 (910B) | **`2.2.rc1`** | `0.12.0`, **`0.11.0`** | `0.5.6.post2` |
|
|
64
|
+
| 8.3 (310P) | **`2.2.rc1`** | | |
|
|
65
|
+
| 8.2 (A3/910C) | **`2.1.rc2`** | ~~`0.11.0`~~, `0.10.2`, <br/>`0.10.1.1` | `0.5.2`, `0.5.1.post3` |
|
|
66
|
+
| 8.2 (910B) | **`2.1.rc2`**, `2.1.rc1` | ~~`0.11.0`~~, `0.10.2`, <br/>`0.10.1.1`, `0.10.0`, <br/>`0.9.2`, `0.9.1` | `0.5.2`, `0.5.1.post3` |
|
|
67
|
+
| 8.2 (310P) | **`2.1.rc2`**, `2.1.rc1` | `0.10.0`, `0.9.2` | |
|
|
67
68
|
|
|
68
69
|
### Iluvatar CoreX
|
|
69
70
|
|
|
@@ -82,14 +83,15 @@ The following table lists the supported accelerated backends and their correspon
|
|
|
82
83
|
`7.5 8.0+PTX 8.9 9.0+PTX`.
|
|
83
84
|
|
|
84
85
|
> [!IMPORTANT]
|
|
85
|
-
> -
|
|
86
|
+
> - Applied [Qwen2.5 VL patched](https://github.com/gpustack/gpustack/issues/3606) to vLLM 0.11.2.
|
|
87
|
+
> - Applied [vLLM[audio] packages](https://github.com/vllm-project/vllm/blob/275de34170654274616082721348b7edd9741d32/setup.py#L720-L724) to vLLM 0.11.2.
|
|
86
88
|
|
|
87
|
-
| CUDA Version <br/> (Variant) | vLLM
|
|
88
|
-
|
|
89
|
-
| 12.9 | `0.12.0`, **`0.11.2`** | `0.5.6.post2` | |
|
|
90
|
-
| 12.8 | `0.12.0`, **`0.11.2`**, <br/>`0.11.0`, `0.10.2`, <br/>`0.10.1.1`, `0.10.0` | `0.5.6.post2`, `0.5.5.post3`, <br/>`0.5.5`, `0.5.4.post3` | `0.0.20` |
|
|
91
|
-
| 12.6 | `0.12.0`, **`0.11.2`**, <br/>`0.11.0`, `0.10.2`, <br/>`0.10.1.1`, `0.10.0` | `0.5.6.post2` | `0.0.20` |
|
|
92
|
-
| 12.4 | `0.11.0`, `0.10.2`, <br/>`0.10.1.1`, `0.10.0`
|
|
89
|
+
| CUDA Version <br/> (Variant) | vLLM | SGLang | VoxBox |
|
|
90
|
+
|------------------------------|--------------------------------------------------------------------------------------|-----------------------------------------------------------|----------|
|
|
91
|
+
| 12.9 | `0.13.0`, `0.12.0`, **`0.11.2`** | `0.5.6.post2` | |
|
|
92
|
+
| 12.8 | `0.13.0`, `0.12.0`, **`0.11.2`**, <br/>`0.11.0`, `0.10.2`, <br/>`0.10.1.1`, `0.10.0` | `0.5.6.post2`, `0.5.5.post3`, <br/>`0.5.5`, `0.5.4.post3` | `0.0.20` |
|
|
93
|
+
| 12.6 | `0.13.0`, `0.12.0`, **`0.11.2`**, <br/>`0.11.0`, `0.10.2`, <br/>`0.10.1.1`, `0.10.0` | `0.5.6.post2` | `0.0.20` |
|
|
94
|
+
| 12.4 | `0.11.0`, `0.10.2`, <br/>`0.10.1.1`, `0.10.0` | | `0.0.20` |
|
|
93
95
|
|
|
94
96
|
### Hygon DTK
|
|
95
97
|
|
|
@@ -113,15 +115,20 @@ The following table lists the supported accelerated backends and their correspon
|
|
|
113
115
|
`gfx908 gfx90a gfx942 gfx1030 gfx1100`.
|
|
114
116
|
|
|
115
117
|
> [!WARNING]
|
|
116
|
-
> - ROCm 7.0 vLLM `0.11.2
|
|
117
|
-
7.0 specific PyTorch build. Although ROCm
|
|
118
|
+
> - ROCm 7.0 vLLM `0.11.2/0.11.0` are reusing the official ROCm 6.4 PyTorch 2.9 wheel package rather than a ROCm
|
|
119
|
+
7.0 specific PyTorch build. Although supports ROCm 7.0 in vLLM `0.11.2/0.11.0`, `gfx1150/gfx1151` are not supported yet.
|
|
118
120
|
> - SGLang supports `gfx942` only.
|
|
121
|
+
> - ROCm 6.4 vLLM `0.13.0` supports `gfx903 gfx90a gfx942` only.
|
|
119
122
|
|
|
120
|
-
|
|
121
|
-
|
|
122
|
-
|
|
123
|
-
|
|
124
|
-
|
|
|
123
|
+
> [!IMPORTANT]
|
|
124
|
+
> - Applied [vLLM[audio] packages](https://github.com/vllm-project/vllm/blob/275de34170654274616082721348b7edd9741d32/setup.py#L720-L724) to vLLM 0.11.2.
|
|
125
|
+
> - Applied [petit-kernel package](https://github.com/vllm-project/vllm/blob/275de34170654274616082721348b7edd9741d32/setup.py#L728) to vLLM 0.11.2 and SGLang 0.5.5.post3.
|
|
126
|
+
|
|
127
|
+
| ROCm Version <br/> (Variant) | vLLM | SGLang |
|
|
128
|
+
|------------------------------|--------------------------------------------------|----------------------------------|
|
|
129
|
+
| 7.0 | `0.13.0`, `0.12.0`, **`0.11.2`**, <br/> `0.11.0` | `0.5.6.post2` |
|
|
130
|
+
| 6.4 | `0.13.0`, `0.12.0`, **`0.11.2`**, <br/> `0.10.2` | `0.5.6.post2`, **`0.5.5.post3`** |
|
|
131
|
+
| 6.3 | `0.10.1.1`, `0.10.0` | |
|
|
125
132
|
|
|
126
133
|
## Directory Structure
|
|
127
134
|
|
|
@@ -1,15 +1,15 @@
|
|
|
1
1
|
gpustack_runner/__init__.py,sha256=0_0jsxo1xjLtHTOIEU0_-A1qFEANzsVw-uXGjcILDwk,530
|
|
2
2
|
gpustack_runner/__main__.py,sha256=wtcp9lwMkaXGbQkuOY08EQhKfIHcTLSjMdnj2W3UGwk,1285
|
|
3
|
-
gpustack_runner/_version.py,sha256=
|
|
3
|
+
gpustack_runner/_version.py,sha256=yMQuvIXXVIrEIokaGWmpbjbyBqb1f3fBWlIy53iYTJs,792
|
|
4
4
|
gpustack_runner/_version.pyi,sha256=A42NoSgcqEXVy2OeNm4LXC9CbyonbooYrSUBlPm2lGY,156
|
|
5
5
|
gpustack_runner/runner.py,sha256=GA7vy4xxaN63cFoc7_Ecft4mV5uYJe_DkQKkR1Htz3o,25881
|
|
6
|
-
gpustack_runner/runner.py.json,sha256=
|
|
6
|
+
gpustack_runner/runner.py.json,sha256=u5qDrFaiWaMgTbT6sS9tUEkI6zbR41yO1PU8QCVYT1o,39041
|
|
7
7
|
gpustack_runner/cmds/__init__.py,sha256=Os8FdvqNjLYiVn_jnDo7rFEtAeVLJJI1odKHEqWF-Fw,417
|
|
8
8
|
gpustack_runner/cmds/__types__.py,sha256=7C4kQM0EHPD8WpJpTo6kh9rEdkrYALcLQ-GAzMMsqV8,789
|
|
9
9
|
gpustack_runner/cmds/images.py,sha256=cFYRCpAdGltYmwkaPmSD7ZFtjlySOTcn4vPTmnUgYwE,38070
|
|
10
|
-
gpustack_runner/_version_appendix.py,sha256=
|
|
11
|
-
gpustack_runner-0.1.22.
|
|
12
|
-
gpustack_runner-0.1.22.
|
|
13
|
-
gpustack_runner-0.1.22.
|
|
14
|
-
gpustack_runner-0.1.22.
|
|
15
|
-
gpustack_runner-0.1.22.
|
|
10
|
+
gpustack_runner/_version_appendix.py,sha256=R9iEnnU0bPP8QLeXZ7sar1vcpPEhr9X3W8R3sQFP-e4,23
|
|
11
|
+
gpustack_runner-0.1.22.post3.dist-info/METADATA,sha256=6iOpRZwepSxVaUf4iIifswJ-OZmJZAwetaK05ZtRM-A,13351
|
|
12
|
+
gpustack_runner-0.1.22.post3.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
|
|
13
|
+
gpustack_runner-0.1.22.post3.dist-info/entry_points.txt,sha256=M1Dxl6cY0kIgf2I4pPsV-_kU6BAtjj93spmsXAdwW3s,66
|
|
14
|
+
gpustack_runner-0.1.22.post3.dist-info/licenses/LICENSE,sha256=OiPibowBvB-NHV3TP_NOj18XNBlXcshXZFMpa3uvKVE,10362
|
|
15
|
+
gpustack_runner-0.1.22.post3.dist-info/RECORD,,
|
|
File without changes
|
{gpustack_runner-0.1.22.post1.dist-info → gpustack_runner-0.1.22.post3.dist-info}/entry_points.txt
RENAMED
|
File without changes
|
{gpustack_runner-0.1.22.post1.dist-info → gpustack_runner-0.1.22.post3.dist-info}/licenses/LICENSE
RENAMED
|
File without changes
|