2026-06-11 06:02:40 UTC
17.7 GB
/usr/local/cuda
CUDA_VERSION12.9.1
DEBIAN_FRONTENDnoninteractive
FLASHINFER_VERSION0.6.11.post1
GDRCOPY_HOME/usr/src/gdrdrv-2.5.1/
LANGen_US.UTF-8
LANGUAGEen_US:en
LC_ALLen_US.UTF-8
LD_LIBRARY_PATH/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
LIBRARY_PATH/usr/local/cuda/lib64/stubs
NCCL_VERSION2.27.3-1
NVARCHx86_64
NVIDIA_DRIVER_CAPABILITIEScompute,utility
NVIDIA_PRODUCT_NAMECUDA
NVIDIA_REQUIRE_CUDAcuda>=12.9 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=vapps,driver>=535,driver<536 brand=vpc,driver>=535,driver<536 brand=vcs,driver>=535,driver<536 brand=vws,driver>=535,driver<536 brand=cloudgaming,driver>=535,driver<536 brand=unknown,driver>=550,driver<551 brand=grid,driver>=550,driver<551 brand=tesla,driver>=550,driver<551 brand=nvidia,driver>=550,driver<551 brand=quadro,driver>=550,driver<551 brand=quadrortx,driver>=550,driver<551 brand=nvidiartx,driver>=550,driver<551 brand=vapps,driver>=550,driver<551 brand=vpc,driver>=550,driver<551 brand=vcs,driver>=550,driver<551 brand=vws,driver>=550,driver<551 brand=cloudgaming,driver>=550,driver<551 brand=unknown,driver>=560,driver<561 brand=grid,driver>=560,driver<561 brand=tesla,driver>=560,driver<561 brand=nvidia,driver>=560,driver<561 brand=quadro,driver>=560,driver<561 brand=quadrortx,driver>=560,driver<561 brand=nvidiartx,driver>=560,driver<561 brand=vapps,driver>=560,driver<561 brand=vpc,driver>=560,driver<561 brand=vcs,driver>=560,driver<561 brand=vws,driver>=560,driver<561 brand=cloudgaming,driver>=560,driver<561 brand=unknown,driver>=565,driver<566 brand=grid,driver>=565,driver<566 brand=tesla,driver>=565,driver<566 brand=nvidia,driver>=565,driver<566 brand=quadro,driver>=565,driver<566 brand=quadrortx,driver>=565,driver<566 brand=nvidiartx,driver>=565,driver<566 brand=vapps,driver>=565,driver<566 brand=vpc,driver>=565,driver<566 brand=vcs,driver>=565,driver<566 brand=vws,driver>=565,driver<566 brand=cloudgaming,driver>=565,driver<566 brand=unknown,driver>=570,driver<571 brand=grid,driver>=570,driver<571 brand=tesla,driver>=570,driver<571 brand=nvidia,driver>=570,driver<571 brand=quadro,driver>=570,driver<571 brand=quadrortx,driver>=570,driver<571 brand=nvidiartx,driver>=570,driver<571 brand=vapps,driver>=570,driver<571 brand=vpc,driver>=570,driver<571 brand=vcs,driver>=570,driver<571 brand=vws,driver>=570,driver<571 brand=cloudgaming,driver>=570,driver<571
NVIDIA_VISIBLE_DEVICESall
NV_CUDA_CUDART_DEV_VERSION12.9.79-1
NV_CUDA_CUDART_VERSION12.9.79-1
NV_CUDA_LIB_VERSION12.9.1-1
NV_CUDA_NSIGHT_COMPUTE_DEV_PACKAGEcuda-nsight-compute-12-9=12.9.1-1
NV_CUDA_NSIGHT_COMPUTE_VERSION12.9.1-1
NV_CUDNN_PACKAGElibcudnn9-cuda-12=9.10.2.21-1
NV_CUDNN_PACKAGE_DEVlibcudnn9-dev-cuda-12=9.10.2.21-1
NV_CUDNN_PACKAGE_DEV_HEADERSlibcudnn9-headers-cuda-12=9.10.2.21-1
NV_CUDNN_PACKAGE_NAMElibcudnn9-cuda-12
NV_CUDNN_VERSION9.10.2.21-1
NV_LIBCUBLAS_DEV_PACKAGElibcublas-dev-12-9=12.9.1.4-1
NV_LIBCUBLAS_DEV_PACKAGE_NAMElibcublas-dev-12-9
NV_LIBCUBLAS_DEV_VERSION12.9.1.4-1
NV_LIBCUBLAS_PACKAGElibcublas-12-9=12.9.1.4-1
NV_LIBCUBLAS_PACKAGE_NAMElibcublas-12-9
NV_LIBCUBLAS_VERSION12.9.1.4-1
NV_LIBCUSPARSE_DEV_VERSION12.5.10.65-1
NV_LIBCUSPARSE_VERSION12.5.10.65-1
NV_LIBNCCL_DEV_PACKAGElibnccl-dev=2.27.3-1+cuda12.9
NV_LIBNCCL_DEV_PACKAGE_NAMElibnccl-dev
NV_LIBNCCL_DEV_PACKAGE_VERSION2.27.3-1
NV_LIBNCCL_PACKAGElibnccl2=2.27.3-1+cuda12.9
NV_LIBNCCL_PACKAGE_NAMElibnccl2
NV_LIBNCCL_PACKAGE_VERSION2.27.3-1
NV_LIBNPP_DEV_PACKAGElibnpp-dev-12-9=12.4.1.87-1
NV_LIBNPP_DEV_VERSION12.4.1.87-1
NV_LIBNPP_PACKAGElibnpp-12-9=12.4.1.87-1
NV_LIBNPP_VERSION12.4.1.87-1
NV_NVML_DEV_VERSION12.9.79-1
NV_NVPROF_DEV_PACKAGEcuda-nvprof-12-9=12.9.79-1
NV_NVPROF_VERSION12.9.79-1
NV_NVTX_VERSION12.9.79-1
OMPI_MCA_coll_hcoll_enable0
OPAL_PREFIX/opt/hpcx/ompi
PATH/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/cuda/nvvm/bin:/usr/local/mpi/bin:/usr/local/ucx/bin:/root/.cargo/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/nvidia/bin:/opt/amazon/efa/bin
PIP_DISABLE_PIP_VERSION_CHECK1
PIP_NO_CACHE_DIR1
PIP_ROOT_USER_ACTIONignore
PYTHON_VERSION3.12
RUNAI_STREAMER_LOG_LEVELINFO
RUNAI_STREAMER_LOG_TO_STDERR1
RUNAI_STREAMER_MEMORY_LIMIT0
SAFETENSORS_FAST_GPU1
SGLANG_ALLOW_OVERWRITE_LONGER_CONTEXT_LEN1
SGLANG_AWS_EFA_VERSION1.46.0
SGLANG_BUILD_COMMIT5a15cde858ea09b77116212a39356f2fc51b8584
SGLANG_BUILD_URLhttps://github.com/sgl-project/sglang/actions/runs/26328908289
SGLANG_DISABLE_CUDNN_CHECK1
SGLANG_IMAGE_TAGlmsysorg/sglang:v0.5.12.post1-cu129
SGLANG_INT4_WEIGHT0
SGLANG_LMCACHE_VERSION0.4.4
SGLANG_MOE_PADDING1
SGLANG_NVIDIA_HPCX_VERSION2.24.1_cuda12
SGLANG_SET_CPU_AFFINITY1
SGLANG_TORCH_CUDA_VERSION12.9.1
SGLANG_TORCH_VERSION2.11.0
SGLANG_VERSION0.5.12.post1
UV_BREAK_SYSTEM_PACKAGES1
UV_HTTP_TIMEOUT500
UV_INDEX_STRATEGYunsafe-best-match
UV_NO_CACHE1
UV_PRERELEASEallow
UV_SKIP_WHEEL_FILENAME_CHECK1
UV_SYSTEM_PYTHON1
[#000] sha256:32f112e3802cadcab3543160f4d2aa607b3cc1c62140d57b4f5441384f40e927 - 0.16% (28.3 MB)
[#001] sha256:644e9b20358325501941bab7efe2465969b1101fd546be263fc7c2d12d2d8c6c - 0.02% (4.34 MB)
[#002] sha256:02559cd4bc8db240554ff3a4e38df5909d4745846b3e3a4df865d102f0a6d0d9 - 0.54% (98.7 MB)
[#003] sha256:2cd52cbb1ebefad9b697ee1c60a7ba7ce8293404a1ad596fd257efdf0e9ec8d6 - 0.0% (182 Bytes)
[#004] sha256:6e8af4fd0a071982e528b634ba99dec2474c21147f99748be708f36e10e3f4c2 - 0.0% (6.72 KB)
[#005] sha256:15a17189b2df6f2a5b84716dd422741e91bec18d40f6cd72f7dea007a02d45aa - 12.1% (2.15 GB)
[#006] sha256:02cb0e091e334f5f0b222d9bc044d174ed386f1b30d3959d0e79ed725eb9cb15 - 0.0% (58.3 KB)
[#007] sha256:9c3d619183d2383e5d618638762511e6711c112a9e8f7e50031f94b839b2c25e - 0.0% (1.64 KB)
[#008] sha256:7f7602a82106bfe07dabfc1cd3799c2b0ee0f1314f4cb28535f7f1b7c00fbec4 - 0.0% (1.49 KB)
[#009] sha256:5a2aba542b08b08acc6ca580eb15aec763e088eed8c62eb5528e0490078951c6 - 16.99% (3.01 GB)
[#010] sha256:6cb9b761b8771e76ec5ff0e90e9165ee60d170c20a9e153b52cef78de6ca15d4 - 0.0% (87.5 KB)
[#011] sha256:bc94a9448b8da7df4edefb8b643b52c0e240a2a3bd97a09b2a45a214fad0cc85 - 3.74% (679 MB)
[#012] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
[#013] sha256:73ca7a4b1bf3cc982ee6ae7cdd7d1f8322b174e9ac226db51481fad91d89fcce - 0.8% (145 MB)
[#014] sha256:ba24ef54577d5dcd932de0623c0865be2fcd6c78a916129388b0de2a540bdbb6 - 2.66% (484 MB)
[#015] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
[#016] sha256:20c6ce7e5044ed944fa14239effb76cbc5b485d9b6970e5d2ae6ae71870bae8d - 0.01% (1.43 MB)
[#017] sha256:e112132ac129b81c2a9942a08856b992235300eedc2c68aeb02e290f2d96f568 - 0.0% (186 Bytes)
[#018] sha256:625fce28c10db26eb1a0eae9eedc7145411a4989f24f119a21def35c892ecf1e - 0.0% (846 KB)
[#019] sha256:787fdcf02bd7486fb9e4f931609fb0ba39c302af24ac2bbe39fa090c574a8625 - 0.0% (103 Bytes)
[#020] sha256:ea4408ace15e9308fa47e50f834ae22911ac3661321601f73dfc5b34dbf2482d - 1.13% (205 MB)
[#021] sha256:7b421f5e0ec62fabb0344ad3a63771eddbd94e76916126e4a34e48bee331024e - 1.83% (333 MB)
[#022] sha256:7c431c0c1b036584c1c2623a062a903548c82389b2df024df18a9cc2fb3d33e3 - 0.0% (2.3 KB)
[#023] sha256:e0215b5f704f2991ea9afa1b837b724c46168bbf30e03eb91718bf8aa30c6337 - 0.0% (2.13 KB)
[#024] sha256:7458fdaeb0e05700bad9f1438b4946e666a3ad5422d19775697939698caf206a - 0.0% (1.78 KB)
[#025] sha256:7642f909caa00d9ddcd0f21314d9c75ed72902c6ebc2e09a34f6a5f20ae1ef34 - 35.74% (6.34 GB)
[#026] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
[#027] sha256:90a78fc9d9d262980265e429f8538a9f551a04cc51247008143d2e65abfbf70b - 0.09% (15.8 MB)
[#028] sha256:ba7e51e64cadb62fc1e57115d575eee2b57e7e584f18e64747709dda602cfbfc - 0.17% (30.3 MB)
[#029] sha256:cc7e9bdc54baf473283b648b3470310c13c1f07bcdd75deb3118839b13689cfa - 0.09% (16 MB)
[#030] sha256:87c17893af5784865a456254a661ee92a30396e8bfa0af144f5cef9d9555df95 - 10.53% (1.87 GB)
[#031] sha256:10be1af09c2d51bce3ce28fe0627d930dbb3546fb8baee17ceb65a9451879642 - 0.0% (13.1 KB)
[#032] sha256:d7d4c32ad8f97fb2ea89e5a22df2af032ae99ec1b05b51fcf2bfbaeb49fc654a - 0.01% (1.7 MB)
[#033] sha256:57639bbe147912ca451d3aa9611413f7faa2ec9b442e1bf1aaab484f31cec4fe - 0.45% (81 MB)
[#034] sha256:6f9f0f71d04522151688ec4098f4962c728922ba92b5e8c745be467cec82d4e0 - 0.01% (957 KB)
[#035] sha256:b6e7de9f3d5f9e25d0d8c0c78013e5b28da66daa6af724d6432c4ae1c1420440 - 0.04% (6.92 MB)
[#036] sha256:f31d76e3b0284e8140339059722283cd3296cdf03914ec890510a6ee36a021bb - 0.04% (7.52 MB)
[#037] sha256:04ed0c7b16b6546699d2583b34eea8414da2d51ea904eaaa47fd63f51955bcdc - 0.04% (7.11 MB)
[#038] sha256:7406050bfef6c1764d26cfca95097c04674fa885801e2bd29388fb6aeb4bedc7 - 0.01% (1.93 MB)
[#039] sha256:d283e5f0972dbfd1ffad87917d7a64a1498e2c19e831faaca9c515537050c55a - 0.01% (1.9 MB)
[#040] sha256:9f373c649d16f3e90cd49d71661738924bf41c5a77116d878c81fcf95cdb5a4c - 0.04% (6.98 MB)
[#041] sha256:a8d02f0068df994af7145eaa70a898f6a22bf075c41ec5e2fd4bfd1bca103011 - 0.35% (63.9 MB)
[#042] sha256:d9f925e936b99351ed9a313090e160f37d6c0a6a5e77dc803861149e162a7e9e - 1.12% (203 MB)
[#043] sha256:536b6889560c1c6016738a2ca06a1e227317c1454d02659146f4f673a7667fe1 - 0.24% (43.3 MB)
[#044] sha256:4e4ac669a24b5473f316a1f0415ab93967dbe33afe5e6c163c86311813fe5fb4 - 0.61% (111 MB)
[#045] sha256:5c80908995bc04636ca52cb2fe3f1cdcef7cc758b8a3326c2eaaeeadb88b83f6 - 0.91% (164 MB)
[#046] sha256:86ff40ae41efa2137947ec5f3a643f1ee868037c49fe26b5f516ce74423fe146 - 0.0% (443 Bytes)
[#047] sha256:1d0b9278f7f4ee6c4069e09e32deae1d56605b81dda0d2a89cecc7fc7cd84210 - 0.0% (723 Bytes)
[#048] sha256:09a2eb838ec1bb11cb8d06a41ede7e13c36cd1ce9d1e3bcc040bdd03b85c617a - 0.0% (507 Bytes)
[#049] sha256:1c6d375ffae4e06b0e1cc681c8c61d7d1d33c72d0bdde7478b875cbb96961ecb - 0.0% (519 Bytes)
[#050] sha256:b46205888920e0c57afd29d7c363fb1eabc7a0eb75eb1f134c5d0441ab7a7d05 - 0.0% (393 Bytes)
[#051] sha256:c3f6df46cc011064f33b9bcd2f074e22392443d672c3f7d3515609eb915ccb57 - 0.0% (421 KB)
[#052] sha256:544acdf14c8d0c6a79e268cb5c38dfc92bd9aa00fc9a92a48e3207539385bf80 - 0.05% (9.49 MB)
[#053] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
[#054] sha256:ac1dba5a37af9f5a79d9e987bae5e6a332b66de89b15e8c06756bcf37c7470c9 - 0.95% (173 MB)
[#055] sha256:59941830f64271589b00608cd29f59d83e2d089b7f96037bd9c55bfb7d4c4c47 - 0.22% (39.6 MB)
[#056] sha256:86ac9b0409e021d6b31f07dfb2a1cbce3620d2286d9039f98e25f06dc1cd3873 - 3.03% (551 MB)
[#057] sha256:120780ec3481bb7faa405d90fb007bab19c78890751077109980cca42f3708b4 - 0.09% (16.9 MB)
[#058] sha256:f128113ac676ce9f4fde49bb43dce4227a4ad9bbd6a680d137affae798a09c6a - 0.17% (30.4 MB)
[#059] sha256:ff7226e50d48d9f72c5c69b8f9a4bc428657fcaac57517401e07630a15ab7ff5 - 0.22% (40.1 MB)
[#060] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
[#061] sha256:04a7ec3a364a1cbd5e55b7693b24581ef7ae076a7b7a5716186ca2995a6f33bd - 0.42% (75.8 MB)
[#062] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
[#063] sha256:3fb4f80aa80d625a86abc53384b5bb49ea71c884f9a0674a2592d9c1496e5103 - 0.18% (33.4 MB)
[#064] sha256:12c78b48be78de917e99241dfe747f94b38665ddf56f3e2af172cd76094e59df - 2.31% (420 MB)
[#065] sha256:77861a3343a5831e50f66be26ada0918842c1e9b62b9da2f14cb3f1d4ee6ceac - 0.07% (13.2 MB)
[#066] sha256:749763e881e53b0408048f5117fc6b983cb82a4f8e9fcaea94ae7fc02a3f7030 - 1.42% (258 MB)
[#067] sha256:a80ba2b1338f28963ed2cffefac05c14eae8283672c7189e94208ddc0344f382 - 0.39% (71.4 MB)
[#068] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
/bin/sh -c #(nop) ARG RELEASE
2025-07-14 16:40:20 UTC/bin/sh -c #(nop) ARG LAUNCHPAD_BUILD_ARCH
2025-07-14 16:40:20 UTC/bin/sh -c #(nop) LABEL org.opencontainers.image.ref.name=ubuntu
2025-07-14 16:40:20 UTC/bin/sh -c #(nop) LABEL org.opencontainers.image.version=24.04
2025-07-14 16:40:22 UTC/bin/sh -c #(nop) ADD file:b4619a63cd7829e1338ddaa4995ca17003002dd54b0dfd675a6f54a2b69151a6 in /
2025-07-14 16:40:23 UTC/bin/sh -c #(nop) CMD ["/bin/bash"]
2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)ENV NVARCH=x86_64
2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)ENV NVIDIA_REQUIRE_CUDA=cuda>=12.9 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=vapps,driver>=535,driver<536 brand=vpc,driver>=535,driver<536 brand=vcs,driver>=535,driver<536 brand=vws,driver>=535,driver<536 brand=cloudgaming,driver>=535,driver<536 brand=unknown,driver>=550,driver<551 brand=grid,driver>=550,driver<551 brand=tesla,driver>=550,driver<551 brand=nvidia,driver>=550,driver<551 brand=quadro,driver>=550,driver<551 brand=quadrortx,driver>=550,driver<551 brand=nvidiartx,driver>=550,driver<551 brand=vapps,driver>=550,driver<551 brand=vpc,driver>=550,driver<551 brand=vcs,driver>=550,driver<551 brand=vws,driver>=550,driver<551 brand=cloudgaming,driver>=550,driver<551 brand=unknown,driver>=560,driver<561 brand=grid,driver>=560,driver<561 brand=tesla,driver>=560,driver<561 brand=nvidia,driver>=560,driver<561 brand=quadro,driver>=560,driver<561 brand=quadrortx,driver>=560,driver<561 brand=nvidiartx,driver>=560,driver<561 brand=vapps,driver>=560,driver<561 brand=vpc,driver>=560,driver<561 brand=vcs,driver>=560,driver<561 brand=vws,driver>=560,driver<561 brand=cloudgaming,driver>=560,driver<561 brand=unknown,driver>=565,driver<566 brand=grid,driver>=565,driver<566 brand=tesla,driver>=565,driver<566 brand=nvidia,driver>=565,driver<566 brand=quadro,driver>=565,driver<566 brand=quadrortx,driver>=565,driver<566 brand=nvidiartx,driver>=565,driver<566 brand=vapps,driver>=565,driver<566 brand=vpc,driver>=565,driver<566 brand=vcs,driver>=565,driver<566 brand=vws,driver>=565,driver<566 brand=cloudgaming,driver>=565,driver<566 brand=unknown,driver>=570,driver<571 brand=grid,driver>=570,driver<571 brand=tesla,driver>=570,driver<571 brand=nvidia,driver>=570,driver<571 brand=quadro,driver>=570,driver<571 brand=quadrortx,driver>=570,driver<571 brand=nvidiartx,driver>=570,driver<571 brand=vapps,driver>=570,driver<571 brand=vpc,driver>=570,driver<571 brand=vcs,driver>=570,driver<571 brand=vws,driver>=570,driver<571 brand=cloudgaming,driver>=570,driver<571
2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)ENV NV_CUDA_CUDART_VERSION=12.9.79-1
2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)ARG TARGETARCH
2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com>
2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends gnupg2 curl ca-certificates && curl -fsSL https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/${NVARCH}/3bf863cc.pub | apt-key add - && echo "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/${NVARCH} /" > /etc/apt/sources.list.d/cuda.list && apt-get purge --autoremove -y curl && rm -rf /var/lib/apt/lists/* # buildkit
2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)ENV CUDA_VERSION=12.9.1
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends cuda-cudart-12-9=${NV_CUDA_CUDART_VERSION} cuda-compat-12-9 && rm -rf /var/lib/apt/lists/* # buildkit
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c echo "/usr/local/cuda/lib64" >> /etc/ld.so.conf.d/nvidia.conf # buildkit
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)ENV PATH=/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)COPY NGC-DL-CONTAINER-LICENSE / # buildkit
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)ENV NVIDIA_VISIBLE_DEVICES=all
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_CUDA_LIB_VERSION=12.9.1-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_NVTX_VERSION=12.9.79-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBNPP_VERSION=12.4.1.87-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBNPP_PACKAGE=libnpp-12-9=12.4.1.87-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBCUSPARSE_VERSION=12.5.10.65-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBCUBLAS_PACKAGE_NAME=libcublas-12-9
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBCUBLAS_VERSION=12.9.1.4-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBCUBLAS_PACKAGE=libcublas-12-9=12.9.1.4-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBNCCL_PACKAGE_NAME=libnccl2
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBNCCL_PACKAGE_VERSION=2.27.3-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NCCL_VERSION=2.27.3-1
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NV_LIBNCCL_PACKAGE=libnccl2=2.27.3-1+cuda12.9
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ARG TARGETARCH
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com>
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends cuda-libraries-12-9=${NV_CUDA_LIB_VERSION} ${NV_LIBNPP_PACKAGE} cuda-nvtx-12-9=${NV_NVTX_VERSION} libcusparse-12-9=${NV_LIBCUSPARSE_VERSION} ${NV_LIBCUBLAS_PACKAGE} ${NV_LIBNCCL_PACKAGE} && rm -rf /var/lib/apt/lists/* # buildkit
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c apt-mark hold ${NV_LIBCUBLAS_PACKAGE_NAME} ${NV_LIBNCCL_PACKAGE_NAME} # buildkit
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)COPY entrypoint.d/ /opt/nvidia/entrypoint.d/ # buildkit
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)COPY nvidia_entrypoint.sh /opt/nvidia/ # buildkit
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENV NVIDIA_PRODUCT_NAME=CUDA
2025-07-18 20:22:35 UTC (buildkit.dockerfile.v0)ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_CUDA_LIB_VERSION=12.9.1-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_CUDA_CUDART_DEV_VERSION=12.9.79-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_NVML_DEV_VERSION=12.9.79-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBCUSPARSE_DEV_VERSION=12.5.10.65-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBNPP_DEV_VERSION=12.4.1.87-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBNPP_DEV_PACKAGE=libnpp-dev-12-9=12.4.1.87-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBCUBLAS_DEV_VERSION=12.9.1.4-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBCUBLAS_DEV_PACKAGE_NAME=libcublas-dev-12-9
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBCUBLAS_DEV_PACKAGE=libcublas-dev-12-9=12.9.1.4-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_CUDA_NSIGHT_COMPUTE_VERSION=12.9.1-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_CUDA_NSIGHT_COMPUTE_DEV_PACKAGE=cuda-nsight-compute-12-9=12.9.1-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_NVPROF_VERSION=12.9.79-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_NVPROF_DEV_PACKAGE=cuda-nvprof-12-9=12.9.79-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBNCCL_DEV_PACKAGE_NAME=libnccl-dev
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBNCCL_DEV_PACKAGE_VERSION=2.27.3-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NCCL_VERSION=2.27.3-1
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV NV_LIBNCCL_DEV_PACKAGE=libnccl-dev=2.27.3-1+cuda12.9
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ARG TARGETARCH
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com>
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends cuda-cudart-dev-12-9=${NV_CUDA_CUDART_DEV_VERSION} cuda-command-line-tools-12-9=${NV_CUDA_LIB_VERSION} cuda-minimal-build-12-9=${NV_CUDA_LIB_VERSION} cuda-libraries-dev-12-9=${NV_CUDA_LIB_VERSION} cuda-nvml-dev-12-9=${NV_NVML_DEV_VERSION} ${NV_NVPROF_DEV_PACKAGE} ${NV_LIBNPP_DEV_PACKAGE} libcusparse-dev-12-9=${NV_LIBCUSPARSE_DEV_VERSION} ${NV_LIBCUBLAS_DEV_PACKAGE} ${NV_LIBNCCL_DEV_PACKAGE} ${NV_CUDA_NSIGHT_COMPUTE_DEV_PACKAGE} && rm -rf /var/lib/apt/lists/* # buildkit
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c apt-mark hold ${NV_LIBCUBLAS_DEV_PACKAGE_NAME} ${NV_LIBNCCL_DEV_PACKAGE_NAME} # buildkit
2025-07-18 20:37:48 UTC (buildkit.dockerfile.v0)ENV LIBRARY_PATH=/usr/local/cuda/lib64/stubs
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)ENV NV_CUDNN_VERSION=9.10.2.21-1
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)ENV NV_CUDNN_PACKAGE_NAME=libcudnn9-cuda-12
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)ENV NV_CUDNN_PACKAGE=libcudnn9-cuda-12=9.10.2.21-1
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)ENV NV_CUDNN_PACKAGE_DEV=libcudnn9-dev-cuda-12=9.10.2.21-1
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)ENV NV_CUDNN_PACKAGE_DEV_HEADERS=libcudnn9-headers-cuda-12=9.10.2.21-1
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)ARG TARGETARCH
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com>
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)LABEL com.nvidia.cudnn.version=9.10.2.21-1
2025-07-25 15:14:12 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends ${NV_CUDNN_PACKAGE} ${NV_CUDNN_PACKAGE_DEV} ${NV_CUDNN_PACKAGE_DEV_HEADERS} && apt-mark hold ${NV_CUDNN_PACKAGE_NAME} && rm -rf /var/lib/apt/lists/* # buildkit
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=amd64
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG BUILD_TYPE=all
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG BRANCH_TYPE=remote
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG GRACE_BLACKWELL=0
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG HOPPER_SBO=0
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG BUILD_AND_DOWNLOAD_PARALLEL=8
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG SGL_KERNEL_VERSION=0.4.2.post2
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG SGL_VERSION=0.5.12.post1
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG SGL_DEEP_GEMM_VERSION=0.1.0
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG USE_LATEST_SGLANG=0
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG GDRCOPY_VERSION=2.5.1
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG PIP_DEFAULT_INDEX
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG UBUNTU_MIRROR
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG GITHUB_ARTIFACTORY=github.com
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG INSTALL_FLASHINFER_JIT_CACHE=1
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG FLASHINFER_VERSION=0.6.11.post1
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG MOONCAKE_VERSION=0.3.10.post2
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ARG MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ENV DEBIAN_FRONTEND=noninteractive CUDA_HOME=/usr/local/cuda GDRCOPY_HOME=/usr/src/gdrdrv-2.5.1/ FLASHINFER_VERSION=0.6.11.post1
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)ENV PATH=/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/nvidia/bin LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
2026-05-23 10:23:55 UTC (buildkit.dockerfile.v0)RUN |20 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON /bin/sh -c if [ -n "$UBUNTU_MIRROR" ]; then sed -i "s|http://.*archive.ubuntu.com|$UBUNTU_MIRROR|g" /etc/apt/sources.list && sed -i "s|http://.*security.ubuntu.com|$UBUNTU_MIRROR|g" /etc/apt/sources.list; fi # buildkit
2026-05-23 10:24:21 UTC (buildkit.dockerfile.v0)RUN |20 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON /bin/sh -c apt update && apt install -y --no-install-recommends wget software-properties-common && apt install -y --no-install-recommends python3.12-full python3.12-dev && update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.12 2 && update-alternatives --set python3 /usr/bin/python3.12 && wget -q https://bootstrap.pypa.io/get-pip.py && python3 get-pip.py --break-system-packages && rm get-pip.py && python3 -m pip config set global.break-system-packages true && cd /usr/lib/python3/dist-packages/ && ln -s apt_pkg.cpython-312-*-linux-gnu.so apt_pkg.so # buildkit
2026-05-23 10:25:21 UTC (buildkit.dockerfile.v0)RUN |20 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON /bin/sh -c apt-get update && apt-get install -y --no-install-recommends ca-certificates software-properties-common netcat-openbsd kmod unzip openssh-server curl wget lsof locales build-essential cmake perl patchelf ccache git-lfs libopenmpi-dev libnuma1 libnuma-dev numactl ffmpeg libibverbs-dev libibverbs1 libibumad3 librdmacm1 libnl-3-200 libnl-route-3-200 libnl-route-3-dev libnl-3-dev ibverbs-providers infiniband-diags perftest libgoogle-glog-dev libgtest-dev libjsoncpp-dev libunwind-dev libboost-all-dev libssl-dev libgrpc-dev libgrpc++-dev libprotobuf-dev protobuf-compiler protobuf-compiler-grpc pybind11-dev libhiredis-dev libcurl4-openssl-dev libczmq4 libczmq-dev libfabric-dev linux-libc-dev devscripts debhelper fakeroot dkms check libsubunit0 libsubunit-dev && ln -sf /usr/bin/python3.12 /usr/bin/python && rm -rf /var/lib/apt/lists/* && apt-get clean # buildkit
2026-05-23 10:25:21 UTC (buildkit.dockerfile.v0)RUN |20 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON /bin/sh -c if [ -n "${PIP_DEFAULT_INDEX}" ]; then python3 -m pip config set global.index-url ${PIP_DEFAULT_INDEX}; fi # buildkit
2026-05-23 10:26:03 UTC (buildkit.dockerfile.v0)RUN |20 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON /bin/sh -c mkdir -p /tmp/gdrcopy && cd /tmp && curl --retry 3 --retry-delay 2 -fsSL -o v${GDRCOPY_VERSION}.tar.gz https://${GITHUB_ARTIFACTORY}/NVIDIA/gdrcopy/archive/refs/tags/v${GDRCOPY_VERSION}.tar.gz && tar -xzf v${GDRCOPY_VERSION}.tar.gz && rm v${GDRCOPY_VERSION}.tar.gz && cd gdrcopy-${GDRCOPY_VERSION}/packages && CUDA=/usr/local/cuda ./build-deb-packages.sh && dpkg -i gdrdrv-dkms_*.deb libgdrapi_*.deb gdrcopy-tests_*.deb gdrcopy_*.deb && cd / && rm -rf /tmp/gdrcopy # buildkit
2026-05-23 10:26:03 UTC (buildkit.dockerfile.v0)RUN |20 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON /bin/sh -c ln -sf /usr/lib/$(uname -m)-linux-gnu/libmlx5.so.1 /usr/lib/$(uname -m)-linux-gnu/libmlx5.so # buildkit
2026-05-23 10:26:04 UTC (buildkit.dockerfile.v0)RUN |20 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON /bin/sh -c locale-gen en_US.UTF-8 # buildkit
2026-05-23 10:26:04 UTC (buildkit.dockerfile.v0)ENV LANG=en_US.UTF-8 LANGUAGE=en_US:en LC_ALL=en_US.UTF-8
2026-05-23 10:26:04 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION=12.9.1
2026-05-23 10:26:04 UTC (buildkit.dockerfile.v0)ARG BUILD_TYPE=all
2026-05-23 10:26:04 UTC (buildkit.dockerfile.v0)ARG SGL_KERNEL_VERSION
2026-05-23 10:26:04 UTC (buildkit.dockerfile.v0)ARG GITHUB_ARTIFACTORY
2026-05-23 10:26:04 UTC (buildkit.dockerfile.v0)WORKDIR /sgl-workspace
2026-05-23 10:26:04 UTC (buildkit.dockerfile.v0)ENV PATH=/root/.cargo/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/nvidia/bin
2026-05-23 10:26:14 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c curl --proto '=https' --tlsv1.2 --retry 3 --retry-delay 2 -sSf https://sh.rustup.rs | sh -s -- -y --no-modify-path --profile minimal && rustc --version && cargo --version # buildkit
2026-05-23 10:26:31 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c python3 -m pip install --upgrade pip setuptools wheel html5lib six && case "$CUDA_VERSION" in 12.6.1) CUINDEX=126 ;; 12.8.1) CUINDEX=128 ;; 12.9.1) CUINDEX=129 ;; 13.0.1) CUINDEX=130 ;; *) echo "Unsupported CUDA version: $CUDA_VERSION" && exit 1 ;; esac && if [ "$CUDA_VERSION" = "12.6.1" ]; then python3 -m pip install https://${GITHUB_ARTIFACTORY}/sgl-project/whl/releases/download/v${SGL_KERNEL_VERSION}/sglang_kernel-${SGL_KERNEL_VERSION}+cu124-cp310-abi3-manylinux2014_$(uname -m).whl --force-reinstall --no-deps ; elif [ "$CUDA_VERSION" = "12.8.1" ] || [ "$CUDA_VERSION" = "12.9.1" ]; then python3 -m pip install https://github.com/sgl-project/whl/releases/download/v${SGL_KERNEL_VERSION}/sglang_kernel-${SGL_KERNEL_VERSION}+cu129-cp310-abi3-manylinux2014_$(uname -m).whl --force-reinstall --no-deps ; elif [ "$CUDA_VERSION" = "13.0.1" ]; then python3 -m pip install sglang-kernel==${SGL_KERNEL_VERSION} --force-reinstall --no-deps ; else echo "Unsupported CUDA version: $CUDA_VERSION" && exit 1 ; fi # buildkit
2026-05-23 10:26:31 UTC (buildkit.dockerfile.v0)COPY python/pyproject.toml /tmp/sglang_deps/python/pyproject.toml # buildkit
2026-05-23 10:26:31 UTC (buildkit.dockerfile.v0)COPY rust/sglang-grpc /tmp/sglang_deps/rust/sglang-grpc # buildkit
2026-05-23 10:26:31 UTC (buildkit.dockerfile.v0)COPY proto /tmp/sglang_deps/proto # buildkit
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c case "$CUDA_VERSION" in 12.6.1) CUINDEX=126 ;; 12.8.1) CUINDEX=128 ;; 12.9.1) CUINDEX=129 ;; 13.0.1) CUINDEX=130 ;; *) echo "Unsupported CUDA version: $CUDA_VERSION" && exit 1 ;; esac && cd /tmp/sglang_deps/python && mkdir -p sglang && touch sglang/__init__.py && echo '__version__ = "0.0.0"' > sglang/version.py && touch README.md && touch LICENSE && python3 -m pip install --extra-index-url https://download.pytorch.org/whl/cu${CUINDEX} ".[${BUILD_TYPE}]" && if [ "${CUDA_VERSION%%.*}" = "12" ]; then pip list --format=freeze | awk -F'==' '/-cu13(==|$)/ {print $1}' | xargs -r python3 -m pip uninstall -y && python3 -m pip install --index-url https://download.pytorch.org/whl/cu${CUINDEX} torch torchvision torchaudio --force-reinstall; python3 -m pip install https://github.com/sgl-project/whl/releases/download/v${SGL_DEEP_GEMM_VERSION}/sgl_deep_gemm-${SGL_DEEP_GEMM_VERSION}+cu129-py3-none-manylinux2014_$(uname -m).whl --force-reinstall; fi && cd /sgl-workspace && rm -rf /tmp/sglang_deps && pip freeze | grep -v "^sglang==" > /sgl-workspace/constraints.txt # buildkit
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG BRANCH_TYPE
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG BUILD_TYPE=all
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION=12.9.1
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG BUILD_AND_DOWNLOAD_PARALLEL
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG SGL_VERSION=0.5.12.post1
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG USE_LATEST_SGLANG
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG GITHUB_ARTIFACTORY
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG MOONCAKE_VERSION
2026-05-23 10:32:17 UTC (buildkit.dockerfile.v0)ARG MOONCAKE_COMPILE_ARG
2026-05-23 10:32:18 UTC (buildkit.dockerfile.v0)WORKDIR /sgl-workspace
2026-05-23 10:35:05 UTC (buildkit.dockerfile.v0)COPY /wheels /tmp/wheels/deepep # buildkit
2026-05-23 10:35:05 UTC (buildkit.dockerfile.v0)COPY /build/DeepEP /sgl-workspace/DeepEP # buildkit
2026-05-23 10:35:06 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c pip install /tmp/wheels/deepep/*.whl && rm -rf /tmp/wheels/deepep # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /flashinfer_jit_output/ /usr/local/lib/python3.12/dist-packages/ # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /tools/diff-so-fancy /usr/local/bin/ # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /tools/clang-format /usr/local/bin/ # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /tools/clangd /usr/local/bin/ # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /tools/lib /usr/local/lib/ # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /tools/cmake /usr/local/bin/ # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /tools/ctest /usr/local/bin/ # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /tools/cpack /usr/local/bin/ # buildkit
2026-05-23 10:35:24 UTC (buildkit.dockerfile.v0)COPY /tools/share/cmake-3.31 /usr/local/share/cmake-3.31 # buildkit
2026-05-23 10:35:25 UTC (buildkit.dockerfile.v0)COPY /tools/just /usr/local/bin/ # buildkit
2026-05-23 10:35:25 UTC (buildkit.dockerfile.v0)COPY /root/.oh-my-zsh /root/.oh-my-zsh # buildkit
2026-05-23 10:35:37 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends gdb ninja-build vim tmux htop zsh tree silversearcher-ag cloc pkg-config bear less rdma-core openssh-server gnuplot infiniband-diags perftest ibverbs-providers libibumad3 libibverbs1 libnl-3-200 libnl-route-3-200 librdmacm1 && rm -rf /var/lib/apt/lists/* && apt-get clean # buildkit
2026-05-23 10:36:02 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c apt update -y && apt install -y --no-install-recommends gnupg && echo "deb http://developer.download.nvidia.com/devtools/repos/ubuntu2004/$(if [ "$(uname -m)" = "aarch64" ]; then echo "arm64"; else echo "amd64"; fi) /" | tee /etc/apt/sources.list.d/nvidia-devtools.list && apt-key adv --fetch-keys http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/$(if [ "$(uname -m)" = "aarch64" ]; then echo "arm64"; else echo "x86_64"; fi)/7fa2af80.pub && apt update -y && apt install -y --no-install-recommends nsight-systems-cli && rm -rf /var/lib/apt/lists/* # buildkit
2026-05-23 10:36:05 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c CUDA_MAJOR="${CUDA_VERSION%%.*}" && if [ "$CUDA_MAJOR" -ge 13 ]; then python3 -m pip install mooncake-transfer-engine-cuda13==${MOONCAKE_VERSION}; else python3 -m pip install mooncake-transfer-engine==${MOONCAKE_VERSION}; fi # buildkit
2026-05-23 10:36:18 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c python3 -m pip install -c /sgl-workspace/constraints.txt datamodel_code_generator pre-commit pytest black isort icdiff uv wheel scikit-build-core py-spy cubloaty google-cloud-storage pandas matplotlib tabulate termplotlib "runai-model-streamer[s3,gcs,azure]>=0.15.7" # buildkit
2026-05-23 10:40:52 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c if [ "${CUDA_VERSION%%.*}" = "12" ]; then python3 -m pip install nixl nixl-cu12 --no-deps ; python3 -m pip install cuda-python==12.9 ; cd /sgl-workspace && git clone https://github.com/deepseek-ai/FlashMLA.git flash-mla && cd flash-mla && git submodule update --init --recursive && pip install --no-build-isolation -v . ; elif [ "${CUDA_VERSION%%.*}" = "13" ]; then python3 -m pip install nixl nixl-cu13 --no-deps ; python3 -m pip install cuda-python==13.2.0 ; cd /sgl-workspace && git clone https://github.com/deepseek-ai/FlashMLA.git flash-mla && ln -s /usr/local/cuda/include/cccl/cuda /usr/local/cuda/include/cuda && cd flash-mla && git submodule update --init --recursive && pip install --no-build-isolation -v . ; fi # buildkit
2026-05-23 10:40:52 UTC (buildkit.dockerfile.v0)COPY --chown=root:root --chmod=755 docker/configs/yank /usr/local/bin/yank # buildkit
2026-05-23 10:40:52 UTC (buildkit.dockerfile.v0)COPY docker/configs/opt/.vimrc /opt/sglang/.vimrc # buildkit
2026-05-23 10:40:52 UTC (buildkit.dockerfile.v0)COPY docker/configs/opt/.tmux.conf /opt/sglang/.tmux.conf # buildkit
2026-05-23 10:40:52 UTC (buildkit.dockerfile.v0)COPY docker/configs/opt/.gitconfig /opt/sglang/.gitconfig # buildkit
2026-05-23 10:40:52 UTC (buildkit.dockerfile.v0)COPY docker/configs/.zshrc /root/.zshrc # buildkit
2026-05-23 10:40:53 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c python3 -m pip install --upgrade "urllib3>=2.6.3" "pillow>=12.1.1" # buildkit
2026-05-23 10:41:02 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c apt-get update && apt-get install -y --only-upgrade binutils binutils-common binutils-x86-64-linux-gnu libbinutils libctf0 libctf-nobfd0 libgprofng0 libsframe1 libgnutls30t64 libpam-modules libpam-modules-bin libpam-runtime libpam0g libsqlite3-0 libtasn1-6 dpkg dpkg-dev libdpkg-perl && rm -rf /var/lib/apt/lists/* # buildkit
2026-05-23 10:41:02 UTC (buildkit.dockerfile.v0)ARG BRANCH_TYPE
2026-05-23 10:41:02 UTC (buildkit.dockerfile.v0)ARG BUILD_TYPE=all
2026-05-23 10:41:02 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION=12.9.1
2026-05-23 10:41:02 UTC (buildkit.dockerfile.v0)ARG SGL_VERSION=0.5.12.post1
2026-05-23 10:41:02 UTC (buildkit.dockerfile.v0)ARG USE_LATEST_SGLANG
2026-05-23 10:41:02 UTC (buildkit.dockerfile.v0)WORKDIR /sgl-workspace
2026-05-23 10:41:03 UTC (buildkit.dockerfile.v0)COPY /src /tmp/local_src # buildkit
2026-05-23 10:41:07 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c if [ "$BRANCH_TYPE" = "local" ]; then cp -r /tmp/local_src /sgl-workspace/sglang; elif [ "$USE_LATEST_SGLANG" = "1" ]; then git clone --depth=1 https://github.com/sgl-project/sglang.git /sgl-workspace/sglang; elif [ -z "$SGL_VERSION" ]; then echo "ERROR: SGL_VERSION must be set when USE_LATEST_SGLANG=0 and BRANCH_TYPE!=local" && exit 1; else git clone --depth=1 --branch v${SGL_VERSION} https://github.com/sgl-project/sglang.git /sgl-workspace/sglang; fi && rm -rf /tmp/local_src # buildkit
2026-05-23 10:41:41 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c cd /sgl-workspace/sglang && python3 -m pip install --no-deps -e "python[${BUILD_TYPE}]" && kernels lock python && ( success=0; if [ "$(uname -m)" = "aarch64" ]; then echo "Skipping kernels-community/sgl-flash-attn3 cubin download on aarch64 (no variants published upstream); kernels will be JIT-compiled at runtime"; success=1; else for i in 1 2 3; do echo "Attempt $i/3: downloading sgl-kernel cubins..." && kernels download python && success=1 && break; echo "sgl-kernel cubin download failed, retrying in 30s..." && sleep 30; done; fi; [ "$success" = "1" ] ) && mkdir -p /root/.cache/huggingface /root/.cache/sglang && ( if [ -f python/kernels.lock ]; then mv python/kernels.lock /root/.cache/sglang/; fi ) && ( find /usr/local/lib/python3.12/dist-packages -type d -name "__pycache__" -exec rm -rf {} + 2>/dev/null || true ) # buildkit
2026-05-23 10:41:42 UTC (buildkit.dockerfile.v0)COPY /build/sgl-model-gateway-bin /usr/local/bin/sgl-model-gateway # buildkit
2026-05-23 10:41:42 UTC (buildkit.dockerfile.v0)COPY /build/gateway_wheels /tmp/gateway_wheels # buildkit
2026-05-23 10:41:47 UTC (buildkit.dockerfile.v0)RUN |21 TARGETARCH=amd64 BUILD_TYPE=all BRANCH_TYPE=remote GRACE_BLACKWELL=0 HOPPER_SBO=0 HOPPER_SBO_DEEPEP_COMMIT=9f2fc4b3182a51044ae7ecb6610f7c9c3258c4d6 DEEPEP_COMMIT=9af0e0d0e74f3577af1979c9b9e1ac2cad0104ee BUILD_AND_DOWNLOAD_PARALLEL=8 SGL_KERNEL_VERSION=0.4.2.post2 SGL_VERSION=0.5.12.post1 SGL_DEEP_GEMM_VERSION=0.1.0 USE_LATEST_SGLANG=0 GDRCOPY_VERSION=2.5.1 PIP_DEFAULT_INDEX= UBUNTU_MIRROR= GITHUB_ARTIFACTORY=github.com INSTALL_FLASHINFER_JIT_CACHE=1 FLASHINFER_VERSION=0.6.11.post1 MOONCAKE_VERSION=0.3.10.post2 MOONCAKE_COMPILE_ARG=-DUSE_HTTP=ON -DUSE_MNNVL=ON -DUSE_CUDA=ON -DWITH_EP=ON CUDA_VERSION=12.9.1 /bin/sh -c python3 -m pip install --force-reinstall /tmp/gateway_wheels/*.whl && rm -rf /tmp/gateway_wheels # buildkit
2026-05-23 10:41:47 UTC (buildkit.dockerfile.v0)WORKDIR /sgl-workspace/sglang
2026-05-23 10:41:47 UTC (buildkit.dockerfile.v0)ARG SGLANG_BUILD_COMMIT=5a15cde858ea09b77116212a39356f2fc51b8584
2026-05-23 10:41:47 UTC (buildkit.dockerfile.v0)ARG SGLANG_BUILD_URL=https://github.com/sgl-project/sglang/actions/runs/26328908289
2026-05-23 10:41:47 UTC (buildkit.dockerfile.v0)ARG SGLANG_IMAGE_TAG=lmsysorg/sglang:v0.5.12.post1-cu129
2026-05-23 10:41:47 UTC (buildkit.dockerfile.v0)ENV SGLANG_BUILD_COMMIT=5a15cde858ea09b77116212a39356f2fc51b8584 SGLANG_BUILD_URL=https://github.com/sgl-project/sglang/actions/runs/26328908289 SGLANG_IMAGE_TAG=lmsysorg/sglang:v0.5.12.post1-cu129
2026-05-23 10:41:47 UTC (buildkit.dockerfile.v0)LABEL org.opencontainers.image.source=https://github.com/sgl-project/sglang org.opencontainers.image.revision=5a15cde858ea09b77116212a39356f2fc51b8584 org.opencontainers.image.version=lmsysorg/sglang:v0.5.12.post1-cu129 org.opencontainers.image.url=https://github.com/sgl-project/sglang/actions/runs/26328908289 ai.sglang.build.commit=5a15cde858ea09b77116212a39356f2fc51b8584 ai.sglang.build.url=https://github.com/sgl-project/sglang/actions/runs/26328908289 ai.sglang.image.tag=lmsysorg/sglang:v0.5.12.post1-cu129
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)SHELL [/bin/bash -eo pipefail -c]
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM=linux/amd64
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)ARG TARGETOS=linux
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=amd64
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)ENV DEBIAN_FRONTEND=noninteractive LANG=en_US.UTF-8 LANGUAGE=en_US:en LC_ALL=en_US.UTF-8
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)RUN |3 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 /bin/bash -eo pipefail -c # Tools # Refresh apt-get update -y && apt-get install -y --no-install-recommends \ software-properties-common apt-transport-https \ ca-certificates gnupg2 lsb-release gnupg-agent \ && apt-get update -y # Install apt-get install -y --no-install-recommends \ ca-certificates build-essential binutils bash openssl \ curl wget aria2 \ git git-lfs \ unzip xz-utils \ tzdata locales \ iproute2 iputils-ping ifstat net-tools dnsutils pciutils ipmitool \ rdma-core rdmacm-utils infiniband-diags \ procps sysstat htop \ tini vim jq bc tree # Update locale localedef -i en_US -c -f UTF-8 -A /usr/share/locale/locale.alias en_US.UTF-8 # Update timezone rm -f /etc/localtime \ && ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime \ && echo "Asia/Shanghai" > /etc/timezone \ && dpkg-reconfigure --frontend noninteractive tzdata # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* \ && rm -rf /var/cache/apt # buildkit
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)ARG PYTHON_VERSION=3.12
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)ENV PYTHON_VERSION=3.12
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)RUN |4 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 /bin/bash -eo pipefail -c # Python # Update alternatives if ! command -v python >/dev/null 2>&1 \ && command -v python${PYTHON_VERSION} >/dev/null 2>&1; then update-alternatives --install /usr/bin/python python $(which python${PYTHON_VERSION}) 1 fi # buildkit
2026-06-11 05:46:23 UTC (buildkit.dockerfile.v0)ENV PIP_NO_CACHE_DIR=1 PIP_DISABLE_PIP_VERSION_CHECK=1 PIP_ROOT_USER_ACTION=ignore
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)RUN |4 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 /bin/bash -eo pipefail -c # Buildkit cat <<EOT >/tmp/requirements.txt build cmake<4 setuptools>=77.0.3,<81.0.0 setuptools_scm>=8 wheel pybind11<3 Cython uv EOT pip install -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION=12.9.1
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG CUDA_ARCHS
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ENV CUDA_HOME=/usr/local/cuda CUDA_VERSION=12.9.1 CUDA_ARCHS=
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ENV UV_SYSTEM_PYTHON=1 UV_PRERELEASE=allow UV_SKIP_WHEEL_FILENAME_CHECK=1 UV_NO_CACHE=1 UV_HTTP_TIMEOUT=500 UV_INDEX_STRATEGY=unsafe-best-match UV_BREAK_SYSTEM_PACKAGES=1
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG SGLANG_VERSION=0.5.12.post1
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG SGLANG_TORCH_VERSION=2.11.0
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG SGLANG_TORCH_CUDA_VERSION=12.9.1
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ENV SGLANG_VERSION=0.5.12.post1 SGLANG_TORCH_VERSION=2.11.0 SGLANG_TORCH_CUDA_VERSION=12.9.1
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)SHELL [/bin/bash -eo pipefail -c]
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM=linux/amd64
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG TARGETOS=linux
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=amd64
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ARG SGLANG_NVIDIA_HPCX_VERSION=2.24.1_cuda12
2026-06-11 05:46:29 UTC (buildkit.dockerfile.v0)ENV SGLANG_NVIDIA_HPCX_VERSION=2.24.1_cuda12
2026-06-11 05:47:26 UTC (buildkit.dockerfile.v0)RUN |10 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= SGLANG_VERSION=0.5.12.post1 SGLANG_TORCH_VERSION=2.11.0 SGLANG_TORCH_CUDA_VERSION=12.9.1 SGLANG_NVIDIA_HPCX_VERSION=2.24.1_cuda12 /bin/bash -eo pipefail -c # NVIDIA HPC-X # Prepare rm -f $(dpkg-query -L libibverbs-dev librdmacm-dev libibumad-dev | grep "\(\.so\|\.a\)$") || true IFS="." read -r CUDA_MAJOR CUDA_MINOR CUDA_PATCH <<< "${CUDA_VERSION}" source /etc/os-release # Get Download Version # If SGLANG_NVIDIA_HPCX_VERSION=2.24.1_cuda13, SGLANG_NVIDIA_HPCX_VERSION_DOWNLOAD=2.24.1 # If SGLANG_NVIDIA_HPCX_VERSION=2.22.1rc4, SGLANG_NVIDIA_HPCX_VERSION_DOWNLOAD=2.22.1 # If SGLANG_NVIDIA_HPCX_VERSION=2.21.3, SGLANG_NVIDIA_HPCX_VERSION_DOWNLOAD=2.21.3 if [[ "${SGLANG_NVIDIA_HPCX_VERSION}" == *"_cuda"* ]]; then SGLANG_NVIDIA_HPCX_VERSION_DOWNLOAD=$(echo "${SGLANG_NVIDIA_HPCX_VERSION}" | sed 's/_cuda.*//') elif [[ "${SGLANG_NVIDIA_HPCX_VERSION}" == *"rc"* ]]; then SGLANG_NVIDIA_HPCX_VERSION_DOWNLOAD=$(echo "${SGLANG_NVIDIA_HPCX_VERSION}" | sed 's/rc.*//') else SGLANG_NVIDIA_HPCX_VERSION_DOWNLOAD=${SGLANG_NVIDIA_HPCX_VERSION} fi # Download mkdir -p /opt/hpcx curl --retry 3 --retry-connrefused -fL "https://content.mellanox.com/hpc/hpc-x/v${SGLANG_NVIDIA_HPCX_VERSION}/hpcx-v${SGLANG_NVIDIA_HPCX_VERSION_DOWNLOAD}-gcc-inbox-${ID}${VERSION_ID}-cuda${CUDA_MAJOR}-$(uname -m).tbz" | tar -jxv -C /opt/hpcx --strip-components 1 # Install ln -sf /opt/hpcx/ompi /usr/local/mpi ln -sf /opt/hpcx/ucx /usr/local/ucx sed -i 's/^\(hwloc_base_binding_policy\) = core$/\1 = none/' /opt/hpcx/ompi/etc/openmpi-mca-params.conf sed -i 's/^\(btl = self\)$/#\1/' /opt/hpcx/ompi/etc/openmpi-mca-params.conf cat <<EOT > /etc/ld.so.conf.d/hpcx.conf /opt/hpcx/clusterkit/lib /opt/hpcx/hcoll/lib /opt/hpcx/nccl_rdma_sharp_plugin/lib /opt/hpcx/ncclnet_plugin/lib /opt/hpcx/ompi/lib /opt/hpcx/sharp/lib /opt/hpcx/ucc/lib /opt/hpcx/ucx/lib EOT # Review ldconfig -v # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* \ && rm -rf /var/cache/apt # buildkit
2026-06-11 05:47:26 UTC (buildkit.dockerfile.v0)ENV PATH=/usr/local/mpi/bin:/usr/local/ucx/bin:/root/.cargo/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/nvidia/bin OPAL_PREFIX=/opt/hpcx/ompi OMPI_MCA_coll_hcoll_enable=0
2026-06-11 05:47:26 UTC (buildkit.dockerfile.v0)ARG SGLANG_AWS_EFA_VERSION=1.46.0
2026-06-11 05:47:26 UTC (buildkit.dockerfile.v0)ENV SGLANG_AWS_EFA_VERSION=1.46.0
2026-06-11 05:49:32 UTC (buildkit.dockerfile.v0)RUN |11 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= SGLANG_VERSION=0.5.12.post1 SGLANG_TORCH_VERSION=2.11.0 SGLANG_TORCH_CUDA_VERSION=12.9.1 SGLANG_NVIDIA_HPCX_VERSION=2.24.1_cuda12 SGLANG_AWS_EFA_VERSION=1.46.0 /bin/bash -eo pipefail -c # AWS EFA # Download curl --retry 3 --retry-connrefused -fL "https://efa-installer.amazonaws.com/aws-efa-installer-${SGLANG_AWS_EFA_VERSION}.tar.gz" | tar -zxv -C /tmp # Install pushd /tmp/aws-efa-installer && \ ./efa_installer.sh -y --skip-kmod # Prepare rm /opt/amazon/efa/lib/libfabric.a || true # Review ldconfig -v # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* \ && rm -rf /var/cache/apt # buildkit
2026-06-11 05:49:32 UTC (buildkit.dockerfile.v0)ENV PATH=/usr/local/mpi/bin:/usr/local/ucx/bin:/root/.cargo/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/nvidia/bin:/opt/amazon/efa/bin
2026-06-11 05:49:32 UTC (buildkit.dockerfile.v0)ARG SGLANG_LMCACHE_VERSION=0.4.4
2026-06-11 05:49:32 UTC (buildkit.dockerfile.v0)ENV SGLANG_LMCACHE_VERSION=0.4.4
2026-06-11 06:02:39 UTC (buildkit.dockerfile.v0)RUN |12 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= SGLANG_VERSION=0.5.12.post1 SGLANG_TORCH_VERSION=2.11.0 SGLANG_TORCH_CUDA_VERSION=12.9.1 SGLANG_NVIDIA_HPCX_VERSION=2.24.1_cuda12 SGLANG_AWS_EFA_VERSION=1.46.0 SGLANG_LMCACHE_VERSION=0.4.4 /bin/bash -eo pipefail -c # LMCache # Install uv pip install --no-build-isolation \ /lmcache/workspace/*.whl # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit
2026-06-11 06:02:40 UTC (buildkit.dockerfile.v0)RUN |12 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= SGLANG_VERSION=0.5.12.post1 SGLANG_TORCH_VERSION=2.11.0 SGLANG_TORCH_CUDA_VERSION=12.9.1 SGLANG_NVIDIA_HPCX_VERSION=2.24.1_cuda12 SGLANG_AWS_EFA_VERSION=1.46.0 SGLANG_LMCACHE_VERSION=0.4.4 /bin/bash -eo pipefail -c # Dependencies # Install cat <<EOT >/tmp/requirements.txt # dockerfile extras accelerate hf_transfer bitsandbytes # audio extras librosa EOT uv pip install \ -r /tmp/requirements.txt # Uninstall ## Remove Run:AI Model Streamer to avoid peak memory usage in model loading. ## Note: 0.5.12 base image still ships runai-model-streamer 0.15.9 via the [runai] extra. uv pip uninstall \ runai-model-streamer || true # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit
2026-06-11 06:02:40 UTC (buildkit.dockerfile.v0)RUN |12 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= SGLANG_VERSION=0.5.12.post1 SGLANG_TORCH_VERSION=2.11.0 SGLANG_TORCH_CUDA_VERSION=12.9.1 SGLANG_NVIDIA_HPCX_VERSION=2.24.1_cuda12 SGLANG_AWS_EFA_VERSION=1.46.0 SGLANG_LMCACHE_VERSION=0.4.4 /bin/bash -eo pipefail -c # Postprocess # Review uv pip tree \ --package sglang \ --package sglang-router \ --package sgl-kernel \ --package flashinfer-python \ --package flash-attn \ --package triton \ --package torch \ --package lmcache \ --package deep-ep \ --package deep-gemm \ --package mooncake-transfer-engine # buildkit
2026-06-11 06:02:40 UTC (buildkit.dockerfile.v0)ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/cuda/nvvm/bin:/usr/local/mpi/bin:/usr/local/ucx/bin:/root/.cargo/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/nvidia/bin:/opt/amazon/efa/bin LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64 SGLANG_DISABLE_CUDNN_CHECK=1 SGLANG_ALLOW_OVERWRITE_LONGER_CONTEXT_LEN=1 SGLANG_INT4_WEIGHT=0 SGLANG_MOE_PADDING=1 SGLANG_SET_CPU_AFFINITY=1 SAFETENSORS_FAST_GPU=1 RUNAI_STREAMER_MEMORY_LIMIT=0 RUNAI_STREAMER_LOG_TO_STDERR=1 RUNAI_STREAMER_LOG_LEVEL=INFO
2026-06-11 06:02:40 UTC (buildkit.dockerfile.v0)WORKDIR /
2026-06-11 06:02:40 UTC (buildkit.dockerfile.v0)ENTRYPOINT ["tini" "--"]
Please be careful as this will not just delete the reference but also the actual content!
For example when you have latest and v1.2.3 both pointing to the same image
the deletion of latest will also permanently remove v1.2.3.