gpustack-runner 0.1.24.post1__tar.gz → 0.1.24.post2__tar.gz

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.
Files changed (135) hide show
  1. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/PKG-INFO +13 -9
  2. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/README.md +12 -8
  3. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/_version.py +2 -2
  4. gpustack_runner-0.1.24.post2/gpustack_runner/_version_appendix.py +1 -0
  5. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/cmds/images.py +2 -2
  6. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/runner.py.json +220 -44
  7. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/cann/Dockerfile +82 -184
  8. gpustack_runner-0.1.24.post2/pack/cann/mindie-atb-models_2.3.0_linux-amd64_py3.11_torch2.1.0-abi0.tar.gz +0 -0
  9. gpustack_runner-0.1.24.post2/pack/cann/mindie-atb-models_2.3.0_linux-arm64_py3.11_torch2.1.0-abi0.tar.gz +0 -0
  10. gpustack_runner-0.1.24.post2/pack/cann/patches/mindie.zip +0 -0
  11. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/maca/Dockerfile +70 -3
  12. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/matrix.yaml +19 -9
  13. gpustack_runner-0.1.24.post2/pack/squash_expand_matrix.sh +86 -0
  14. gpustack_runner-0.1.24.post2/pack/squash_image.sh +103 -0
  15. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/test_list_runners_by_backend.json +220 -44
  16. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/test_list_service_runners.json +6 -6
  17. gpustack_runner-0.1.24.post1/gpustack_runner/_version_appendix.py +0 -1
  18. gpustack_runner-0.1.24.post1/pack/cann/mindie-atb-models_2.2.rc1_linux-amd64_py3.11_torch2.1.0-abi0.tar.gz +0 -0
  19. gpustack_runner-0.1.24.post1/pack/cann/mindie-atb-models_2.2.rc1_linux-arm64_py3.11_torch2.1.0-abi0.tar.gz +0 -0
  20. gpustack_runner-0.1.24.post1/pack/cann/patches/mindie.zip +0 -0
  21. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/.codespelldict +0 -0
  22. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/.codespellrc +0 -0
  23. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/.gitattributes +0 -0
  24. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/.gitignore +0 -0
  25. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/.pre-commit-config.yaml +0 -0
  26. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/.python-version +0 -0
  27. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/LICENSE +0 -0
  28. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/Makefile +0 -0
  29. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/docs/index.md +0 -0
  30. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/docs/modules/gpustack_runner.md +0 -0
  31. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/__init__.py +0 -0
  32. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/__main__.py +0 -0
  33. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/__utils__.py +0 -0
  34. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/_version.pyi +0 -0
  35. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/cmds/__init__.py +0 -0
  36. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/cmds/__types__.py +0 -0
  37. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/envs.py +0 -0
  38. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/gpustack_runner/runner.py +0 -0
  39. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/hatch.toml +0 -0
  40. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/mkdocs.yml +0 -0
  41. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251020_vllm_install_lmcache/cann/Dockerfile +0 -0
  42. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251020_vllm_install_lmcache/cuda/Dockerfile +0 -0
  43. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251020_vllm_install_lmcache/matrix.yaml +0 -0
  44. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251020_vllm_install_lmcache/rocm/Dockerfile +0 -0
  45. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251022_vllm_install_ray_client/cann/Dockerfile +0 -0
  46. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251022_vllm_install_ray_client/cuda/Dockerfile +0 -0
  47. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251022_vllm_install_ray_client/matrix.yaml +0 -0
  48. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251022_vllm_install_ray_client/rocm/Dockerfile +0 -0
  49. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251022_vllm_install_ray_default/cuda/Dockerfile +0 -0
  50. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251022_vllm_install_ray_default/matrix.yaml +0 -0
  51. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251022_vllm_install_ray_default/rocm/Dockerfile +0 -0
  52. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251024_vllm_install_nvidia_hpcx/cuda/Dockerfile +0 -0
  53. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251024_vllm_install_nvidia_hpcx/matrix.yaml +0 -0
  54. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251024_vllm_reinstall_lmcache/cuda/Dockerfile +0 -0
  55. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251024_vllm_reinstall_lmcache/matrix.yaml +0 -0
  56. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251029_vllm_reinstall_ray/cann/Dockerfile +0 -0
  57. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251029_vllm_reinstall_ray/matrix.yaml +0 -0
  58. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251103_mindie_refresh_entrypoint/cann/Dockerfile +0 -0
  59. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251103_mindie_refresh_entrypoint/matrix.yaml +0 -0
  60. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251105_vllm_polish_nvidia_hpcx/cuda/Dockerfile +0 -0
  61. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251105_vllm_polish_nvidia_hpcx/matrix.yaml +0 -0
  62. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251106_vllm_install_ep_kernel/cuda/Dockerfile +0 -0
  63. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251106_vllm_install_ep_kernel/matrix.yaml +0 -0
  64. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251107_vllm_reinstall_lmcache/cuda/Dockerfile +0 -0
  65. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251107_vllm_reinstall_lmcache/matrix.yaml +0 -0
  66. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251110_sglang_install_diffusion/cuda/Dockerfile +0 -0
  67. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251110_sglang_install_diffusion/matrix.yaml +0 -0
  68. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251110_sglang_install_flashattn/cuda/Dockerfile +0 -0
  69. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251110_sglang_install_flashattn/matrix.yaml +0 -0
  70. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251125_mindie_install_posix_ipc/cann/Dockerfile +0 -0
  71. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251125_mindie_install_posix_ipc/matrix.yaml +0 -0
  72. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251201_vllm_patch_qwen2_5_vl/cuda/Dockerfile +0 -0
  73. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251201_vllm_patch_qwen2_5_vl/cuda/patches/vllm_001_disable_flashatten_in_qwen2_5_vl.patch +0 -0
  74. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251201_vllm_patch_qwen2_5_vl/matrix.yaml +0 -0
  75. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251209_mindie_install_av/cann/Dockerfile +0 -0
  76. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251209_mindie_install_av/matrix.yaml +0 -0
  77. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251213_mindie_patch_minicpm_qwen2_v2/cann/Dockerfile +0 -0
  78. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251213_mindie_patch_minicpm_qwen2_v2/cann/patches.zip +0 -0
  79. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251213_mindie_patch_minicpm_qwen2_v2/matrix.yaml +0 -0
  80. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251213_sglang_patch_server_args/cuda/Dockerfile +0 -0
  81. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251213_sglang_patch_server_args/cuda/patches/sglang_001_fix_server_args.patch +0 -0
  82. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251213_sglang_patch_server_args/matrix.yaml +0 -0
  83. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251214_cuda_several_patches/cuda/Dockerfile +0 -0
  84. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251214_cuda_several_patches/matrix.yaml +0 -0
  85. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251215_cann_several_patches/cann/Dockerfile +0 -0
  86. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251215_cann_several_patches/matrix.yaml +0 -0
  87. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251216_sglang_uninstall_runai_model_streamer/cuda/Dockerfile +0 -0
  88. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251216_sglang_uninstall_runai_model_streamer/matrix.yaml +0 -0
  89. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251219_rocm_install_petit_kernel/matrix.yaml +0 -0
  90. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251219_rocm_install_petit_kernel/rocm/Dockerfile +0 -0
  91. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251219_vllm_install_audio_extra/cuda/Dockerfile +0 -0
  92. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251219_vllm_install_audio_extra/matrix.yaml +0 -0
  93. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251219_vllm_install_audio_extra/rocm/Dockerfile +0 -0
  94. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251224_mindie_patch_atb_config/cann/Dockerfile +0 -0
  95. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20251224_mindie_patch_atb_config/matrix.yaml +0 -0
  96. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20260105_vllm_install_omni/cann/Dockerfile +0 -0
  97. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20260105_vllm_install_omni/cuda/Dockerfile +0 -0
  98. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20260105_vllm_install_omni/matrix.yaml +0 -0
  99. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/20260105_vllm_install_omni/rocm/Dockerfile +0 -0
  100. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/.post_operation/README.md +0 -0
  101. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/corex/Dockerfile +0 -0
  102. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/cuda/Dockerfile +0 -0
  103. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/discard_runner.sh +0 -0
  104. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/dtk/Dockerfile +0 -0
  105. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/expand_matrix.sh +0 -0
  106. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/hggc/Dockerfile +0 -0
  107. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/merge_runner.sh +0 -0
  108. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/musa/Dockerfile +0 -0
  109. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/prune_runner.sh +0 -0
  110. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/rocm/Dockerfile +0 -0
  111. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pack/rocm/patches/sglang_001_wrong_vram.patch +0 -0
  112. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pyproject.toml +0 -0
  113. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/pytest.ini +0 -0
  114. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/ruff.toml +0 -0
  115. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/__init__.py +0 -0
  116. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/test_docker_image.json +0 -0
  117. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/test_list_backend_runners.json +0 -0
  118. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/test_list_runners_by_prefix.json +0 -0
  119. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/test_merge_image.json +0 -0
  120. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/test_replace_image_with.json +0 -0
  121. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/fixtures/test_split_image.json +0 -0
  122. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/test_runner.py +0 -0
  123. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tests/gpustack_runner/test_utils.py +0 -0
  124. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/activate +0 -0
  125. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/chat.sh +0 -0
  126. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/chat_tool_current_date_time.sh +0 -0
  127. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/chat_tool_get_temperature.sh +0 -0
  128. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/chat_tool_get_weather.sh +0 -0
  129. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/chat_tool_square_of_number.sh +0 -0
  130. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/chat_tool_square_root_of_number.sh +0 -0
  131. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/chat_tool_where_am_i.sh +0 -0
  132. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/run_runner.sh +0 -0
  133. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/tools/run_runner_cluster.sh +0 -0
  134. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/uv.lock +0 -0
  135. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post2}/uv.toml +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: gpustack-runner
3
- Version: 0.1.24.post1
3
+ Version: 0.1.24.post2
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
@@ -54,8 +54,11 @@ The following table lists the supported accelerated backends and their correspon
54
54
 
55
55
  | CANN Version <br/> (Variant) | MindIE | vLLM | SGLang |
56
56
  |------------------------------|-----------|------------------------------------------------------------|------------------------|
57
- | 8.3 (A3/910C) | `2.2.rc1` | `0.13.0`, `0.12.0`, `0.11.0` | `0.5.7`, `0.5.6.post2` |
58
- | 8.3 (910B) | `2.2.rc1` | `0.13.0`, `0.12.0`, `0.11.0` | `0.5.7`, `0.5.6.post2` |
57
+ | 8.5 (A3/910C) | `2.3.0` | `0.14.1`, `0.13.0` | `0.5.8` |
58
+ | 8.5 (910B) | `2.3.0` | `0.14.1`, `0.13.0` | `0.5.8` |
59
+ | 8.5 (310P) | `2.3.0` | `0.14.1` | |
60
+ | 8.3 (A3/910C) | `2.2.rc1` | `0.12.0`, `0.11.0` | `0.5.7`, `0.5.6.post2` |
61
+ | 8.3 (910B) | `2.2.rc1` | `0.12.0`, `0.11.0` | `0.5.7`, `0.5.6.post2` |
59
62
  | 8.3 (310P) | `2.2.rc1` | | |
60
63
  | 8.2 (A3/910C) | `2.1.rc2` | `0.10.2`, `0.10.1.1` | `0.5.2`, `0.5.1.post3` |
61
64
  | 8.2 (910B) | `2.1.rc2` | `0.10.2`, `0.10.1.1`, <br/>`0.10.0`, `0.9.2`, <br/>`0.9.1` | `0.5.2`, `0.5.1.post3` |
@@ -89,7 +92,7 @@ The following table lists the supported accelerated backends and their correspon
89
92
  |-----------------------------|----------------------------|
90
93
  | 25.04 | `0.11.0`, `0.9.2`, `0.8.5` |
91
94
 
92
- ### THead HGGC
95
+ ### T-Head HGGC
93
96
 
94
97
  | HGGC Version <br/> (Variant) | vLLM | SGLang |
95
98
  |------------------------------|----------|---------|
@@ -97,10 +100,11 @@ The following table lists the supported accelerated backends and their correspon
97
100
 
98
101
  ### MetaX MACA
99
102
 
100
- | MACA Version <br/> (Variant) | vLLM |
101
- |------------------------------|----------|
102
- | 3.2 | `0.10.2` |
103
- | 3.0 | `0.9.1` |
103
+ | MACA Version <br/> (Variant) | vLLM | SGLang |
104
+ |------------------------------|----------|---------|
105
+ | 3.3 | `0.11.2` | `0.5.6` |
106
+ | 3.2 | `0.10.2` | |
107
+ | 3.0 | `0.9.1` | |
104
108
 
105
109
  ### MThreads MUSA
106
110
 
@@ -118,7 +122,7 @@ The following table lists the supported accelerated backends and their correspon
118
122
  `gfx908 gfx90a gfx942 gfx1030 gfx1100`.
119
123
 
120
124
  > [!WARNING]
121
- > - 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
125
+ > - ROCm 7.0 vLLM `0.11.2` are reusing the official ROCm 6.4 PyTorch 2.9 wheel package rather than a ROCm
122
126
  7.0 specific PyTorch build. Although supports ROCm 7.0 in vLLM `0.11.2`, `gfx1150/gfx1151` are not supported yet.
123
127
  > - ROCm 6.4 vLLM `0.13.0` supports `gfx903 gfx90a gfx942` only.
124
128
  > - ROCm 6.4 SGLang supports `gfx942` only.
@@ -34,8 +34,11 @@ The following table lists the supported accelerated backends and their correspon
34
34
 
35
35
  | CANN Version <br/> (Variant) | MindIE | vLLM | SGLang |
36
36
  |------------------------------|-----------|------------------------------------------------------------|------------------------|
37
- | 8.3 (A3/910C) | `2.2.rc1` | `0.13.0`, `0.12.0`, `0.11.0` | `0.5.7`, `0.5.6.post2` |
38
- | 8.3 (910B) | `2.2.rc1` | `0.13.0`, `0.12.0`, `0.11.0` | `0.5.7`, `0.5.6.post2` |
37
+ | 8.5 (A3/910C) | `2.3.0` | `0.14.1`, `0.13.0` | `0.5.8` |
38
+ | 8.5 (910B) | `2.3.0` | `0.14.1`, `0.13.0` | `0.5.8` |
39
+ | 8.5 (310P) | `2.3.0` | `0.14.1` | |
40
+ | 8.3 (A3/910C) | `2.2.rc1` | `0.12.0`, `0.11.0` | `0.5.7`, `0.5.6.post2` |
41
+ | 8.3 (910B) | `2.2.rc1` | `0.12.0`, `0.11.0` | `0.5.7`, `0.5.6.post2` |
39
42
  | 8.3 (310P) | `2.2.rc1` | | |
40
43
  | 8.2 (A3/910C) | `2.1.rc2` | `0.10.2`, `0.10.1.1` | `0.5.2`, `0.5.1.post3` |
41
44
  | 8.2 (910B) | `2.1.rc2` | `0.10.2`, `0.10.1.1`, <br/>`0.10.0`, `0.9.2`, <br/>`0.9.1` | `0.5.2`, `0.5.1.post3` |
@@ -69,7 +72,7 @@ The following table lists the supported accelerated backends and their correspon
69
72
  |-----------------------------|----------------------------|
70
73
  | 25.04 | `0.11.0`, `0.9.2`, `0.8.5` |
71
74
 
72
- ### THead HGGC
75
+ ### T-Head HGGC
73
76
 
74
77
  | HGGC Version <br/> (Variant) | vLLM | SGLang |
75
78
  |------------------------------|----------|---------|
@@ -77,10 +80,11 @@ The following table lists the supported accelerated backends and their correspon
77
80
 
78
81
  ### MetaX MACA
79
82
 
80
- | MACA Version <br/> (Variant) | vLLM |
81
- |------------------------------|----------|
82
- | 3.2 | `0.10.2` |
83
- | 3.0 | `0.9.1` |
83
+ | MACA Version <br/> (Variant) | vLLM | SGLang |
84
+ |------------------------------|----------|---------|
85
+ | 3.3 | `0.11.2` | `0.5.6` |
86
+ | 3.2 | `0.10.2` | |
87
+ | 3.0 | `0.9.1` | |
84
88
 
85
89
  ### MThreads MUSA
86
90
 
@@ -98,7 +102,7 @@ The following table lists the supported accelerated backends and their correspon
98
102
  `gfx908 gfx90a gfx942 gfx1030 gfx1100`.
99
103
 
100
104
  > [!WARNING]
101
- > - 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
105
+ > - ROCm 7.0 vLLM `0.11.2` are reusing the official ROCm 6.4 PyTorch 2.9 wheel package rather than a ROCm
102
106
  7.0 specific PyTorch build. Although supports ROCm 7.0 in vLLM `0.11.2`, `gfx1150/gfx1151` are not supported yet.
103
107
  > - ROCm 6.4 vLLM `0.13.0` supports `gfx903 gfx90a gfx942` only.
104
108
  > - ROCm 6.4 SGLang supports `gfx942` only.
@@ -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 = '0.1.24.post1'
31
- __version_tuple__ = version_tuple = (0, 1, 24, 'post1')
30
+ __version__ = version = '0.1.24.post2'
31
+ __version_tuple__ = version_tuple = (0, 1, 24, 'post2')
32
32
  try:
33
33
  from ._version_appendix import git_commit
34
34
  __commit_id__ = commit_id = git_commit
@@ -0,0 +1 @@
1
+ git_commit = "62d75c6"
@@ -464,7 +464,7 @@ class SaveImagesSubCommand(SubCommand):
464
464
  # Submit tasks
465
465
  for task_name, src_img, dst_file, dst_file_relative in saving_tasks:
466
466
  if dst_file.exists():
467
- print(f"{dst_file.name} already exists, skipping save {src_img}.")
467
+ print(f"{dst_file} already exists, skipping save {src_img}.")
468
468
  continue
469
469
  dst_file.parent.mkdir(parents=True, exist_ok=True, mode=0o744)
470
470
 
@@ -976,7 +976,7 @@ class LoadImagesSubCommand(SubCommand):
976
976
  load_parser.add_argument(
977
977
  "input",
978
978
  nargs=OPTIONAL,
979
- help="Input directory to load images (default: current working directory)",
979
+ help="Input directory to load images from (default: current working directory)",
980
980
  )
981
981
 
982
982
  load_parser.set_defaults(func=LoadImagesSubCommand)
@@ -1,35 +1,79 @@
1
1
  [
2
2
  {
3
3
  "backend": "cann",
4
- "backend_version": "8.3",
5
- "original_backend_version": "8.3.rc2",
4
+ "backend_version": "8.5",
5
+ "original_backend_version": "8.5.0",
6
6
  "backend_variant": "a3",
7
7
  "service": "mindie",
8
- "service_version": "2.2.rc1",
8
+ "service_version": "2.3.0",
9
9
  "platform": "linux/amd64",
10
- "docker_image": "gpustack/runner:cann8.3-a3-mindie2.2.rc1",
10
+ "docker_image": "gpustack/runner:cann8.5-a3-mindie2.3.0",
11
11
  "deprecated": false
12
12
  },
13
13
  {
14
14
  "backend": "cann",
15
- "backend_version": "8.3",
16
- "original_backend_version": "8.3.rc2",
15
+ "backend_version": "8.5",
16
+ "original_backend_version": "8.5.0",
17
17
  "backend_variant": "a3",
18
18
  "service": "mindie",
19
- "service_version": "2.2.rc1",
19
+ "service_version": "2.3.0",
20
20
  "platform": "linux/arm64",
21
- "docker_image": "gpustack/runner:cann8.3-a3-mindie2.2.rc1",
21
+ "docker_image": "gpustack/runner:cann8.5-a3-mindie2.3.0",
22
22
  "deprecated": false
23
23
  },
24
24
  {
25
25
  "backend": "cann",
26
- "backend_version": "8.3",
27
- "original_backend_version": "8.3.rc2",
26
+ "backend_version": "8.5",
27
+ "original_backend_version": "8.5.0",
28
28
  "backend_variant": "a3",
29
29
  "service": "sglang",
30
- "service_version": "0.5.7",
30
+ "service_version": "0.5.8",
31
31
  "platform": "linux/arm64",
32
- "docker_image": "gpustack/runner:cann8.3-a3-sglang0.5.7",
32
+ "docker_image": "gpustack/runner:cann8.5-a3-sglang0.5.8",
33
+ "deprecated": false
34
+ },
35
+ {
36
+ "backend": "cann",
37
+ "backend_version": "8.5",
38
+ "original_backend_version": "8.5.0",
39
+ "backend_variant": "a3",
40
+ "service": "vllm",
41
+ "service_version": "0.14.1",
42
+ "platform": "linux/amd64",
43
+ "docker_image": "gpustack/runner:cann8.5-a3-vllm0.14.1",
44
+ "deprecated": false
45
+ },
46
+ {
47
+ "backend": "cann",
48
+ "backend_version": "8.5",
49
+ "original_backend_version": "8.5.0",
50
+ "backend_variant": "a3",
51
+ "service": "vllm",
52
+ "service_version": "0.14.1",
53
+ "platform": "linux/arm64",
54
+ "docker_image": "gpustack/runner:cann8.5-a3-vllm0.14.1",
55
+ "deprecated": false
56
+ },
57
+ {
58
+ "backend": "cann",
59
+ "backend_version": "8.5",
60
+ "original_backend_version": "8.5.0",
61
+ "backend_variant": "a3",
62
+ "service": "vllm",
63
+ "service_version": "0.13.0",
64
+ "platform": "linux/amd64",
65
+ "docker_image": "gpustack/runner:cann8.5-a3-vllm0.13.0",
66
+ "deprecated": false
67
+ },
68
+ {
69
+ "backend": "cann",
70
+ "backend_version": "8.5",
71
+ "original_backend_version": "8.5.0",
72
+ "backend_variant": "a3",
73
+ "service": "vllm",
74
+ "service_version": "0.13.0",
75
+ "platform": "linux/arm64",
76
+ "docker_image": "gpustack/runner:cann8.5-a3-vllm0.13.0",
33
77
  "deprecated": false
34
78
  },
35
79
  {
@@ -37,10 +81,21 @@
37
81
  "backend_version": "8.3",
38
82
  "original_backend_version": "8.3.rc2",
39
83
  "backend_variant": "a3",
40
- "service": "sglang",
41
- "service_version": "0.5.6.post2",
84
+ "service": "mindie",
85
+ "service_version": "2.2.rc1",
86
+ "platform": "linux/amd64",
87
+ "docker_image": "gpustack/runner:cann8.3-a3-mindie2.2.rc1",
88
+ "deprecated": false
89
+ },
90
+ {
91
+ "backend": "cann",
92
+ "backend_version": "8.3",
93
+ "original_backend_version": "8.3.rc2",
94
+ "backend_variant": "a3",
95
+ "service": "mindie",
96
+ "service_version": "2.2.rc1",
42
97
  "platform": "linux/arm64",
43
- "docker_image": "gpustack/runner:cann8.3-a3-sglang0.5.6.post2",
98
+ "docker_image": "gpustack/runner:cann8.3-a3-mindie2.2.rc1",
44
99
  "deprecated": false
45
100
  },
46
101
  {
@@ -48,10 +103,10 @@
48
103
  "backend_version": "8.3",
49
104
  "original_backend_version": "8.3.rc2",
50
105
  "backend_variant": "a3",
51
- "service": "vllm",
52
- "service_version": "0.13.0",
53
- "platform": "linux/amd64",
54
- "docker_image": "gpustack/runner:cann8.3-a3-vllm0.13.0",
106
+ "service": "sglang",
107
+ "service_version": "0.5.7",
108
+ "platform": "linux/arm64",
109
+ "docker_image": "gpustack/runner:cann8.3-a3-sglang0.5.7",
55
110
  "deprecated": false
56
111
  },
57
112
  {
@@ -59,10 +114,10 @@
59
114
  "backend_version": "8.3",
60
115
  "original_backend_version": "8.3.rc2",
61
116
  "backend_variant": "a3",
62
- "service": "vllm",
63
- "service_version": "0.13.0",
117
+ "service": "sglang",
118
+ "service_version": "0.5.6.post2",
64
119
  "platform": "linux/arm64",
65
- "docker_image": "gpustack/runner:cann8.3-a3-vllm0.13.0",
120
+ "docker_image": "gpustack/runner:cann8.3-a3-sglang0.5.6.post2",
66
121
  "deprecated": false
67
122
  },
68
123
  {
@@ -221,35 +276,79 @@
221
276
  },
222
277
  {
223
278
  "backend": "cann",
224
- "backend_version": "8.3",
225
- "original_backend_version": "8.3.rc2",
279
+ "backend_version": "8.5",
280
+ "original_backend_version": "8.5.0",
226
281
  "backend_variant": "910b",
227
282
  "service": "mindie",
228
- "service_version": "2.2.rc1",
283
+ "service_version": "2.3.0",
229
284
  "platform": "linux/amd64",
230
- "docker_image": "gpustack/runner:cann8.3-910b-mindie2.2.rc1",
285
+ "docker_image": "gpustack/runner:cann8.5-910b-mindie2.3.0",
231
286
  "deprecated": false
232
287
  },
233
288
  {
234
289
  "backend": "cann",
235
- "backend_version": "8.3",
236
- "original_backend_version": "8.3.rc2",
290
+ "backend_version": "8.5",
291
+ "original_backend_version": "8.5.0",
237
292
  "backend_variant": "910b",
238
293
  "service": "mindie",
239
- "service_version": "2.2.rc1",
294
+ "service_version": "2.3.0",
240
295
  "platform": "linux/arm64",
241
- "docker_image": "gpustack/runner:cann8.3-910b-mindie2.2.rc1",
296
+ "docker_image": "gpustack/runner:cann8.5-910b-mindie2.3.0",
242
297
  "deprecated": false
243
298
  },
244
299
  {
245
300
  "backend": "cann",
246
- "backend_version": "8.3",
247
- "original_backend_version": "8.3.rc2",
301
+ "backend_version": "8.5",
302
+ "original_backend_version": "8.5.0",
248
303
  "backend_variant": "910b",
249
304
  "service": "sglang",
250
- "service_version": "0.5.7",
305
+ "service_version": "0.5.8",
251
306
  "platform": "linux/arm64",
252
- "docker_image": "gpustack/runner:cann8.3-910b-sglang0.5.7",
307
+ "docker_image": "gpustack/runner:cann8.5-910b-sglang0.5.8",
308
+ "deprecated": false
309
+ },
310
+ {
311
+ "backend": "cann",
312
+ "backend_version": "8.5",
313
+ "original_backend_version": "8.5.0",
314
+ "backend_variant": "910b",
315
+ "service": "vllm",
316
+ "service_version": "0.14.1",
317
+ "platform": "linux/amd64",
318
+ "docker_image": "gpustack/runner:cann8.5-910b-vllm0.14.1",
319
+ "deprecated": false
320
+ },
321
+ {
322
+ "backend": "cann",
323
+ "backend_version": "8.5",
324
+ "original_backend_version": "8.5.0",
325
+ "backend_variant": "910b",
326
+ "service": "vllm",
327
+ "service_version": "0.14.1",
328
+ "platform": "linux/arm64",
329
+ "docker_image": "gpustack/runner:cann8.5-910b-vllm0.14.1",
330
+ "deprecated": false
331
+ },
332
+ {
333
+ "backend": "cann",
334
+ "backend_version": "8.5",
335
+ "original_backend_version": "8.5.0",
336
+ "backend_variant": "910b",
337
+ "service": "vllm",
338
+ "service_version": "0.13.0",
339
+ "platform": "linux/amd64",
340
+ "docker_image": "gpustack/runner:cann8.5-910b-vllm0.13.0",
341
+ "deprecated": false
342
+ },
343
+ {
344
+ "backend": "cann",
345
+ "backend_version": "8.5",
346
+ "original_backend_version": "8.5.0",
347
+ "backend_variant": "910b",
348
+ "service": "vllm",
349
+ "service_version": "0.13.0",
350
+ "platform": "linux/arm64",
351
+ "docker_image": "gpustack/runner:cann8.5-910b-vllm0.13.0",
253
352
  "deprecated": false
254
353
  },
255
354
  {
@@ -257,10 +356,21 @@
257
356
  "backend_version": "8.3",
258
357
  "original_backend_version": "8.3.rc2",
259
358
  "backend_variant": "910b",
260
- "service": "sglang",
261
- "service_version": "0.5.6.post2",
359
+ "service": "mindie",
360
+ "service_version": "2.2.rc1",
361
+ "platform": "linux/amd64",
362
+ "docker_image": "gpustack/runner:cann8.3-910b-mindie2.2.rc1",
363
+ "deprecated": false
364
+ },
365
+ {
366
+ "backend": "cann",
367
+ "backend_version": "8.3",
368
+ "original_backend_version": "8.3.rc2",
369
+ "backend_variant": "910b",
370
+ "service": "mindie",
371
+ "service_version": "2.2.rc1",
262
372
  "platform": "linux/arm64",
263
- "docker_image": "gpustack/runner:cann8.3-910b-sglang0.5.6.post2",
373
+ "docker_image": "gpustack/runner:cann8.3-910b-mindie2.2.rc1",
264
374
  "deprecated": false
265
375
  },
266
376
  {
@@ -268,10 +378,10 @@
268
378
  "backend_version": "8.3",
269
379
  "original_backend_version": "8.3.rc2",
270
380
  "backend_variant": "910b",
271
- "service": "vllm",
272
- "service_version": "0.13.0",
273
- "platform": "linux/amd64",
274
- "docker_image": "gpustack/runner:cann8.3-910b-vllm0.13.0",
381
+ "service": "sglang",
382
+ "service_version": "0.5.7",
383
+ "platform": "linux/arm64",
384
+ "docker_image": "gpustack/runner:cann8.3-910b-sglang0.5.7",
275
385
  "deprecated": false
276
386
  },
277
387
  {
@@ -279,10 +389,10 @@
279
389
  "backend_version": "8.3",
280
390
  "original_backend_version": "8.3.rc2",
281
391
  "backend_variant": "910b",
282
- "service": "vllm",
283
- "service_version": "0.13.0",
392
+ "service": "sglang",
393
+ "service_version": "0.5.6.post2",
284
394
  "platform": "linux/arm64",
285
- "docker_image": "gpustack/runner:cann8.3-910b-vllm0.13.0",
395
+ "docker_image": "gpustack/runner:cann8.3-910b-sglang0.5.6.post2",
286
396
  "deprecated": false
287
397
  },
288
398
  {
@@ -527,6 +637,50 @@
527
637
  "docker_image": "gpustack/runner:cann8.2-910b-vllm0.9.1",
528
638
  "deprecated": false
529
639
  },
640
+ {
641
+ "backend": "cann",
642
+ "backend_version": "8.5",
643
+ "original_backend_version": "8.5.0",
644
+ "backend_variant": "310p",
645
+ "service": "mindie",
646
+ "service_version": "2.3.0",
647
+ "platform": "linux/amd64",
648
+ "docker_image": "gpustack/runner:cann8.5-310p-mindie2.3.0",
649
+ "deprecated": false
650
+ },
651
+ {
652
+ "backend": "cann",
653
+ "backend_version": "8.5",
654
+ "original_backend_version": "8.5.0",
655
+ "backend_variant": "310p",
656
+ "service": "mindie",
657
+ "service_version": "2.3.0",
658
+ "platform": "linux/arm64",
659
+ "docker_image": "gpustack/runner:cann8.5-310p-mindie2.3.0",
660
+ "deprecated": false
661
+ },
662
+ {
663
+ "backend": "cann",
664
+ "backend_version": "8.5",
665
+ "original_backend_version": "8.5.0",
666
+ "backend_variant": "310p",
667
+ "service": "vllm",
668
+ "service_version": "0.14.1",
669
+ "platform": "linux/amd64",
670
+ "docker_image": "gpustack/runner:cann8.5-310p-vllm0.14.1",
671
+ "deprecated": false
672
+ },
673
+ {
674
+ "backend": "cann",
675
+ "backend_version": "8.5",
676
+ "original_backend_version": "8.5.0",
677
+ "backend_variant": "310p",
678
+ "service": "vllm",
679
+ "service_version": "0.14.1",
680
+ "platform": "linux/arm64",
681
+ "docker_image": "gpustack/runner:cann8.5-310p-vllm0.14.1",
682
+ "deprecated": false
683
+ },
530
684
  {
531
685
  "backend": "cann",
532
686
  "backend_version": "8.3",
@@ -1418,6 +1572,28 @@
1418
1572
  "docker_image": "gpustack/runner:hggc12.3-vllm0.11.1",
1419
1573
  "deprecated": false
1420
1574
  },
1575
+ {
1576
+ "backend": "maca",
1577
+ "backend_version": "3.3",
1578
+ "original_backend_version": "3.3.0",
1579
+ "backend_variant": "",
1580
+ "service": "sglang",
1581
+ "service_version": "0.5.6",
1582
+ "platform": "linux/amd64",
1583
+ "docker_image": "gpustack/runner:maca3.3-sglang0.5.6",
1584
+ "deprecated": false
1585
+ },
1586
+ {
1587
+ "backend": "maca",
1588
+ "backend_version": "3.3",
1589
+ "original_backend_version": "3.3.0",
1590
+ "backend_variant": "",
1591
+ "service": "vllm",
1592
+ "service_version": "0.11.2",
1593
+ "platform": "linux/amd64",
1594
+ "docker_image": "gpustack/runner:maca3.3-vllm0.11.2",
1595
+ "deprecated": false
1596
+ },
1421
1597
  {
1422
1598
  "backend": "maca",
1423
1599
  "backend_version": "3.2",