gpustack-runner 0.1.24.post1__tar.gz → 0.1.24.post3__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 (145) hide show
  1. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/PKG-INFO +28 -24
  2. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/README.md +27 -23
  3. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/_version.py +2 -2
  4. gpustack_runner-0.1.24.post3/gpustack_runner/_version_appendix.py +1 -0
  5. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/cmds/images.py +2 -2
  6. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/runner.py.json +335 -49
  7. gpustack_runner-0.1.24.post3/pack/.post_operation/20260129_sglang_reinstall_kernel/cann/Dockerfile +74 -0
  8. gpustack_runner-0.1.24.post3/pack/.post_operation/20260129_sglang_reinstall_kernel/matrix.yaml +28 -0
  9. gpustack_runner-0.1.24.post3/pack/.post_operation/20260129_vllm_patch_dp/cuda/Dockerfile +25 -0
  10. gpustack_runner-0.1.24.post3/pack/.post_operation/20260129_vllm_patch_dp/cuda/patches/vllm_001_wrong_dp_ray.patch +41 -0
  11. gpustack_runner-0.1.24.post3/pack/.post_operation/20260129_vllm_patch_dp/matrix.yaml +55 -0
  12. gpustack_runner-0.1.24.post3/pack/.post_operation/20260129_vllm_patch_dp/rocm/Dockerfile +25 -0
  13. gpustack_runner-0.1.24.post3/pack/.post_operation/20260129_vllm_patch_dp/rocm/patches/vllm_001_wrong_dp_ray.patch +41 -0
  14. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/README.md +2 -0
  15. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/cann/Dockerfile +94 -181
  16. gpustack_runner-0.1.24.post3/pack/cann/mindie-atb-models_2.3.0_linux-amd64_py3.11_torch2.1.0-abi0.tar.gz +0 -0
  17. gpustack_runner-0.1.24.post3/pack/cann/mindie-atb-models_2.3.0_linux-arm64_py3.11_torch2.1.0-abi0.tar.gz +0 -0
  18. gpustack_runner-0.1.24.post3/pack/cann/patches/mindie.zip +0 -0
  19. gpustack_runner-0.1.24.post3/pack/cann/patches/vllm_001_wrong_dp_ray.patch +41 -0
  20. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/cuda/Dockerfile +23 -10
  21. gpustack_runner-0.1.24.post3/pack/cuda/patches/vllm_001_wrong_dp_ray.patch +41 -0
  22. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/maca/Dockerfile +70 -3
  23. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/matrix.yaml +36 -26
  24. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/rocm/Dockerfile +22 -14
  25. gpustack_runner-0.1.24.post3/pack/rocm/patches/vllm_001_wrong_dp_ray.patch +41 -0
  26. gpustack_runner-0.1.24.post3/pack/squash_expand_matrix.sh +86 -0
  27. gpustack_runner-0.1.24.post3/pack/squash_image.sh +103 -0
  28. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/test_list_runners_by_backend.json +335 -49
  29. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/test_list_service_runners.json +6 -6
  30. gpustack_runner-0.1.24.post1/gpustack_runner/_version_appendix.py +0 -1
  31. 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
  32. 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
  33. gpustack_runner-0.1.24.post1/pack/cann/patches/mindie.zip +0 -0
  34. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/.codespelldict +0 -0
  35. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/.codespellrc +0 -0
  36. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/.gitattributes +0 -0
  37. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/.gitignore +0 -0
  38. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/.pre-commit-config.yaml +0 -0
  39. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/.python-version +0 -0
  40. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/LICENSE +0 -0
  41. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/Makefile +0 -0
  42. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/docs/index.md +0 -0
  43. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/docs/modules/gpustack_runner.md +0 -0
  44. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/__init__.py +0 -0
  45. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/__main__.py +0 -0
  46. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/__utils__.py +0 -0
  47. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/_version.pyi +0 -0
  48. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/cmds/__init__.py +0 -0
  49. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/cmds/__types__.py +0 -0
  50. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/envs.py +0 -0
  51. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/gpustack_runner/runner.py +0 -0
  52. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/hatch.toml +0 -0
  53. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/mkdocs.yml +0 -0
  54. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251020_vllm_install_lmcache/cann/Dockerfile +0 -0
  55. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251020_vllm_install_lmcache/cuda/Dockerfile +0 -0
  56. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251020_vllm_install_lmcache/matrix.yaml +0 -0
  57. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251020_vllm_install_lmcache/rocm/Dockerfile +0 -0
  58. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251022_vllm_install_ray_client/cann/Dockerfile +0 -0
  59. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251022_vllm_install_ray_client/cuda/Dockerfile +0 -0
  60. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251022_vllm_install_ray_client/matrix.yaml +0 -0
  61. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251022_vllm_install_ray_client/rocm/Dockerfile +0 -0
  62. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251022_vllm_install_ray_default/cuda/Dockerfile +0 -0
  63. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251022_vllm_install_ray_default/matrix.yaml +0 -0
  64. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251022_vllm_install_ray_default/rocm/Dockerfile +0 -0
  65. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251024_vllm_install_nvidia_hpcx/cuda/Dockerfile +0 -0
  66. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251024_vllm_install_nvidia_hpcx/matrix.yaml +0 -0
  67. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251024_vllm_reinstall_lmcache/cuda/Dockerfile +0 -0
  68. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251024_vllm_reinstall_lmcache/matrix.yaml +0 -0
  69. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251029_vllm_reinstall_ray/cann/Dockerfile +0 -0
  70. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251029_vllm_reinstall_ray/matrix.yaml +0 -0
  71. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251103_mindie_refresh_entrypoint/cann/Dockerfile +0 -0
  72. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251103_mindie_refresh_entrypoint/matrix.yaml +0 -0
  73. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251105_vllm_polish_nvidia_hpcx/cuda/Dockerfile +0 -0
  74. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251105_vllm_polish_nvidia_hpcx/matrix.yaml +0 -0
  75. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251106_vllm_install_ep_kernel/cuda/Dockerfile +0 -0
  76. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251106_vllm_install_ep_kernel/matrix.yaml +0 -0
  77. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251107_vllm_reinstall_lmcache/cuda/Dockerfile +0 -0
  78. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251107_vllm_reinstall_lmcache/matrix.yaml +0 -0
  79. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251110_sglang_install_diffusion/cuda/Dockerfile +0 -0
  80. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251110_sglang_install_diffusion/matrix.yaml +0 -0
  81. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251110_sglang_install_flashattn/cuda/Dockerfile +0 -0
  82. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251110_sglang_install_flashattn/matrix.yaml +0 -0
  83. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251125_mindie_install_posix_ipc/cann/Dockerfile +0 -0
  84. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251125_mindie_install_posix_ipc/matrix.yaml +0 -0
  85. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251201_vllm_patch_qwen2_5_vl/cuda/Dockerfile +0 -0
  86. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251201_vllm_patch_qwen2_5_vl/cuda/patches/vllm_001_disable_flashatten_in_qwen2_5_vl.patch +0 -0
  87. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251201_vllm_patch_qwen2_5_vl/matrix.yaml +0 -0
  88. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251209_mindie_install_av/cann/Dockerfile +0 -0
  89. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251209_mindie_install_av/matrix.yaml +0 -0
  90. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251213_mindie_patch_minicpm_qwen2_v2/cann/Dockerfile +0 -0
  91. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251213_mindie_patch_minicpm_qwen2_v2/cann/patches.zip +0 -0
  92. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251213_mindie_patch_minicpm_qwen2_v2/matrix.yaml +0 -0
  93. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251213_sglang_patch_server_args/cuda/Dockerfile +0 -0
  94. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251213_sglang_patch_server_args/cuda/patches/sglang_001_fix_server_args.patch +0 -0
  95. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251213_sglang_patch_server_args/matrix.yaml +0 -0
  96. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251214_cuda_several_patches/cuda/Dockerfile +0 -0
  97. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251214_cuda_several_patches/matrix.yaml +0 -0
  98. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251215_cann_several_patches/cann/Dockerfile +0 -0
  99. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251215_cann_several_patches/matrix.yaml +0 -0
  100. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251216_sglang_uninstall_runai_model_streamer/cuda/Dockerfile +0 -0
  101. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251216_sglang_uninstall_runai_model_streamer/matrix.yaml +0 -0
  102. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251219_rocm_install_petit_kernel/matrix.yaml +0 -0
  103. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251219_rocm_install_petit_kernel/rocm/Dockerfile +0 -0
  104. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251219_vllm_install_audio_extra/cuda/Dockerfile +0 -0
  105. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251219_vllm_install_audio_extra/matrix.yaml +0 -0
  106. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251219_vllm_install_audio_extra/rocm/Dockerfile +0 -0
  107. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251224_mindie_patch_atb_config/cann/Dockerfile +0 -0
  108. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20251224_mindie_patch_atb_config/matrix.yaml +0 -0
  109. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20260105_vllm_install_omni/cann/Dockerfile +0 -0
  110. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20260105_vllm_install_omni/cuda/Dockerfile +0 -0
  111. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20260105_vllm_install_omni/matrix.yaml +0 -0
  112. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/.post_operation/20260105_vllm_install_omni/rocm/Dockerfile +0 -0
  113. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/corex/Dockerfile +0 -0
  114. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/discard_runner.sh +0 -0
  115. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/dtk/Dockerfile +0 -0
  116. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/expand_matrix.sh +0 -0
  117. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/hggc/Dockerfile +0 -0
  118. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/merge_runner.sh +0 -0
  119. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/musa/Dockerfile +0 -0
  120. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/prune_runner.sh +0 -0
  121. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pack/rocm/patches/sglang_001_wrong_vram.patch +0 -0
  122. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pyproject.toml +0 -0
  123. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/pytest.ini +0 -0
  124. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/ruff.toml +0 -0
  125. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/__init__.py +0 -0
  126. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/test_docker_image.json +0 -0
  127. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/test_list_backend_runners.json +0 -0
  128. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/test_list_runners_by_prefix.json +0 -0
  129. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/test_merge_image.json +0 -0
  130. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/test_replace_image_with.json +0 -0
  131. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/fixtures/test_split_image.json +0 -0
  132. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/test_runner.py +0 -0
  133. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tests/gpustack_runner/test_utils.py +0 -0
  134. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/activate +0 -0
  135. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/chat.sh +0 -0
  136. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/chat_tool_current_date_time.sh +0 -0
  137. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/chat_tool_get_temperature.sh +0 -0
  138. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/chat_tool_get_weather.sh +0 -0
  139. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/chat_tool_square_of_number.sh +0 -0
  140. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/chat_tool_square_root_of_number.sh +0 -0
  141. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/chat_tool_where_am_i.sh +0 -0
  142. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/run_runner.sh +0 -0
  143. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/tools/run_runner_cluster.sh +0 -0
  144. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/uv.lock +0 -0
  145. {gpustack_runner-0.1.24.post1 → gpustack_runner-0.1.24.post3}/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.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
@@ -52,14 +52,17 @@ The following table lists the supported accelerated backends and their correspon
52
52
  vllm-ascend [#3316](https://github.com/vllm-project/vllm-ascend/issues/3316)
53
53
  and [#2795](https://github.com/vllm-project/vllm-ascend/issues/2795).
54
54
 
55
- | CANN Version <br/> (Variant) | MindIE | vLLM | SGLang |
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` |
59
- | 8.3 (310P) | `2.2.rc1` | | |
60
- | 8.2 (A3/910C) | `2.1.rc2` | `0.10.2`, `0.10.1.1` | `0.5.2`, `0.5.1.post3` |
61
- | 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` |
62
- | 8.2 (310P) | `2.1.rc2` | `0.10.0`, `0.9.2` | |
55
+ | CANN Version <br/> (Variant) | MindIE | vLLM | SGLang |
56
+ |------------------------------|-----------|--------------------------------------------------------------------|------------------------|
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` |
62
+ | 8.3 (310P) | `2.2.rc1` | | |
63
+ | 8.2 (A3/910C) | `2.1.rc2` | `0.10.2`, ~~`0.10.1.1`~~ | `0.5.2`, `0.5.1.post3` |
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` |
65
+ | 8.2 (310P) | `2.1.rc2` | `0.10.0`, `0.9.2` | |
63
66
 
64
67
  ### Iluvatar CoreX
65
68
 
@@ -77,11 +80,11 @@ The following table lists the supported accelerated backends and their correspon
77
80
  > - CUDA 12.6/12.4 supports Compute Capabilities:
78
81
  `7.5 8.0+PTX 8.9 9.0+PTX`.
79
82
 
80
- | CUDA Version <br/> (Variant) | vLLM | SGLang | VoxBox |
81
- |------------------------------|---------------------------------------------|-----------------------------------------------------------|----------|
82
- | 12.9 | `0.13.0`, `0.12.0`, <br/>`0.11.2` | `0.5.7`, `0.5.6.post2` | |
83
- | 12.8 | `0.13.0`, `0.12.0`, <br/>`0.11.2`, `0.10.2` | `0.5.7`, `0.5.6.post2`, <br/>`0.5.5.post3`, `0.5.4.post3` | `0.0.21` |
84
- | 12.6 | `0.13.0`, `0.12.0`, <br/>`0.11.2`,`0.10.2` | | `0.0.21` |
83
+ | CUDA Version <br/> (Variant) | vLLM | SGLang | VoxBox |
84
+ |------------------------------|----------------------------------------------------------------|-----------------------------------------------------------------------------|----------|
85
+ | 12.9 | `0.14.1`, **`0.13.0`**, <br/>`0.12.0`, `0.11.2` | `0.5.8`, `0.5.7`, <br/>`0.5.6.post2` | |
86
+ | 12.8 | `0.14.1`, **`0.13.0`**, <br/>`0.12.0`, `0.11.2`, <br/>`0.10.2` | `0.5.8`, `0.5.7`, <br/>`0.5.6.post2`, `0.5.5.post3`, <br/>~~`0.5.4.post3`~~ | `0.0.21` |
87
+ | 12.6 | `0.14.1`, **`0.13.0`**, <br/>`0.12.0`, `0.11.2`, <br/>`0.10.2` | | `0.0.21` |
85
88
 
86
89
  ### Hygon DTK
87
90
 
@@ -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
 
@@ -118,16 +122,16 @@ 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.
125
129
  > - ROCm 7.0 SGLang supports `gfx950` only.
126
130
 
127
- | ROCm Version <br/> (Variant) | vLLM | SGLang |
128
- |------------------------------|---------------------------------------------|--------------------------------------------|
129
- | 7.0 | `0.13.0`, `0.12.0`, <br/>`0.11.2` | `0.5.7`, `0.5.6.post2` |
130
- | 6.4 | `0.13.0`, `0.12.0`, <br/>`0.11.2`, `0.10.2` | `0.5.7`, `0.5.6.post2`, <br/>`0.5.5.post3` |
131
+ | ROCm Version <br/> (Variant) | vLLM | SGLang |
132
+ |------------------------------|-------------------------------------------------|--------------------------------------------|
133
+ | 7.0 | **`0.13.0`**, `0.12.0`, <br/>`0.11.2` | `0.5.7`, `0.5.6.post2` |
134
+ | 6.4 | **`0.13.0`**, `0.12.0`, <br/>`0.11.2`, `0.10.2` | `0.5.7`, `0.5.6.post2`, <br/>`0.5.5.post3` |
131
135
 
132
136
  ## Directory Structure
133
137
 
@@ -32,14 +32,17 @@ The following table lists the supported accelerated backends and their correspon
32
32
  vllm-ascend [#3316](https://github.com/vllm-project/vllm-ascend/issues/3316)
33
33
  and [#2795](https://github.com/vllm-project/vllm-ascend/issues/2795).
34
34
 
35
- | CANN Version <br/> (Variant) | MindIE | vLLM | SGLang |
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` |
39
- | 8.3 (310P) | `2.2.rc1` | | |
40
- | 8.2 (A3/910C) | `2.1.rc2` | `0.10.2`, `0.10.1.1` | `0.5.2`, `0.5.1.post3` |
41
- | 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` |
42
- | 8.2 (310P) | `2.1.rc2` | `0.10.0`, `0.9.2` | |
35
+ | CANN Version <br/> (Variant) | MindIE | vLLM | SGLang |
36
+ |------------------------------|-----------|--------------------------------------------------------------------|------------------------|
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` |
42
+ | 8.3 (310P) | `2.2.rc1` | | |
43
+ | 8.2 (A3/910C) | `2.1.rc2` | `0.10.2`, ~~`0.10.1.1`~~ | `0.5.2`, `0.5.1.post3` |
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` |
45
+ | 8.2 (310P) | `2.1.rc2` | `0.10.0`, `0.9.2` | |
43
46
 
44
47
  ### Iluvatar CoreX
45
48
 
@@ -57,11 +60,11 @@ The following table lists the supported accelerated backends and their correspon
57
60
  > - CUDA 12.6/12.4 supports Compute Capabilities:
58
61
  `7.5 8.0+PTX 8.9 9.0+PTX`.
59
62
 
60
- | CUDA Version <br/> (Variant) | vLLM | SGLang | VoxBox |
61
- |------------------------------|---------------------------------------------|-----------------------------------------------------------|----------|
62
- | 12.9 | `0.13.0`, `0.12.0`, <br/>`0.11.2` | `0.5.7`, `0.5.6.post2` | |
63
- | 12.8 | `0.13.0`, `0.12.0`, <br/>`0.11.2`, `0.10.2` | `0.5.7`, `0.5.6.post2`, <br/>`0.5.5.post3`, `0.5.4.post3` | `0.0.21` |
64
- | 12.6 | `0.13.0`, `0.12.0`, <br/>`0.11.2`,`0.10.2` | | `0.0.21` |
63
+ | CUDA Version <br/> (Variant) | vLLM | SGLang | VoxBox |
64
+ |------------------------------|----------------------------------------------------------------|-----------------------------------------------------------------------------|----------|
65
+ | 12.9 | `0.14.1`, **`0.13.0`**, <br/>`0.12.0`, `0.11.2` | `0.5.8`, `0.5.7`, <br/>`0.5.6.post2` | |
66
+ | 12.8 | `0.14.1`, **`0.13.0`**, <br/>`0.12.0`, `0.11.2`, <br/>`0.10.2` | `0.5.8`, `0.5.7`, <br/>`0.5.6.post2`, `0.5.5.post3`, <br/>~~`0.5.4.post3`~~ | `0.0.21` |
67
+ | 12.6 | `0.14.1`, **`0.13.0`**, <br/>`0.12.0`, `0.11.2`, <br/>`0.10.2` | | `0.0.21` |
65
68
 
66
69
  ### Hygon DTK
67
70
 
@@ -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
 
@@ -98,16 +102,16 @@ 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.
105
109
  > - ROCm 7.0 SGLang supports `gfx950` only.
106
110
 
107
- | ROCm Version <br/> (Variant) | vLLM | SGLang |
108
- |------------------------------|---------------------------------------------|--------------------------------------------|
109
- | 7.0 | `0.13.0`, `0.12.0`, <br/>`0.11.2` | `0.5.7`, `0.5.6.post2` |
110
- | 6.4 | `0.13.0`, `0.12.0`, <br/>`0.11.2`, `0.10.2` | `0.5.7`, `0.5.6.post2`, <br/>`0.5.5.post3` |
111
+ | ROCm Version <br/> (Variant) | vLLM | SGLang |
112
+ |------------------------------|-------------------------------------------------|--------------------------------------------|
113
+ | 7.0 | **`0.13.0`**, `0.12.0`, <br/>`0.11.2` | `0.5.7`, `0.5.6.post2` |
114
+ | 6.4 | **`0.13.0`**, `0.12.0`, <br/>`0.11.2`, `0.10.2` | `0.5.7`, `0.5.6.post2`, <br/>`0.5.5.post3` |
111
115
 
112
116
  ## Directory Structure
113
117
 
@@ -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.post3'
31
+ __version_tuple__ = version_tuple = (0, 1, 24, 'post3')
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 = "dc41ed2"
@@ -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)