Commit Graph

569 Commits

Author SHA1 Message Date
Anant Sharma 6608cd5947
fix: stage sccache off PATH to prevent meson auto detection (#6578)
Signed-off-by: Anant Sharma <anants@nvidia.com>
2026-02-25 16:01:44 -05:00
Yuewei Na 8483e4a07f
chore: Upgrade to tensorrt-llm==1.3.0rc5 (#6579)
Signed-off-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: Yuewei Na <nv-yna@users.noreply.github.com>
2026-02-25 12:22:58 -08:00
Anant Sharma 73c02a0b4a
fix: disable media-ffmpeg feature by default (#6574)
Signed-off-by: Anant Sharma <anants@nvidia.com>
2026-02-25 11:49:58 -05:00
Ran Rubin 8fbe722895
chore: bake sccache into dynamo base (#6532) 2026-02-25 09:50:07 +02:00
Jason Zhou 4ea80e9585
chore: use aic release/0.7.0 (#6494) 2026-02-24 16:27:19 -06:00
ishandhanani 6642e23e0f
feat: sglang to 0.5.9 + updated docs (#6518)
Co-authored-by: baihuitian <baihuitian.bht@gmail.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-24 21:48:25 +00:00
Keiven C f89088007e
fix: vLLM Dockerfile need nvlink and crt for kvbm dependencies (#6521)
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-24 13:28:44 -08:00
Ayush Agarwal e94f464760
chore: install vllm-omni in vllm container (#6458)
Signed-off-by: ayushag <ayushag@nvidia.com>
2026-02-24 21:03:53 +00:00
Anant Sharma 0f47842ac7
ci: switch container-validation-dynamo to runtime target (#6499)
Signed-off-by: Anant Sharma <anants@nvidia.com>
2026-02-24 09:42:41 -05:00
Keiven C 6d3e0137cf
fix: add missing nvlink binary to vLLM and TRT-LLM runtime/dev/local-dev Dockerfiles (#6503)
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-23 18:33:19 -08:00
William Zhang a798e08c81
fix: Fix dev container build (#6455)
There were 2 bugs prior to this commit when building a TRTLLM dev
container:

- template rendering bug
- CUDA compatibility layer

The `nvidia_entrypoint.sh` script should render the second completely
unnecessary.
2026-02-20 11:49:04 -08:00
KrishnanPrash 0688d5843a
docs: Fix Dockerfile path in README for vLLM dev image (#6429)
Signed-off-by: KrishnanPrash <140860868+KrishnanPrash@users.noreply.github.com>
2026-02-19 11:34:58 -08:00
Yuewei Na 9a15730a7f
chore: Upgrade to tensorrt-llm==1.3.0rc3 (#6402)
Signed-off-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: Yuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: dagil-nvidia <dagil@nvidia.com>
2026-02-19 14:12:32 -05:00
Tushar Sharma 741c20bf6c
fix: move ffmpeg copy before USER dynamo to fix permission denied (#6260)
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Co-authored-by: Dmitry Tokarev <dtokarev@nvidia.com>
2026-02-19 17:45:36 +00:00
Anant Sharma 72762dac00
ci: migrate container-validation-dynamo to self-hosted runners (#6381)
Signed-off-by: Anant Sharma <anants@nvidia.com>
2026-02-19 09:26:04 -05:00
Ran Rubin af17da3504
ci: remote builders sccache support (#6369) 2026-02-18 22:31:26 +02:00
Ran Rubin d38954c759
chore: optimize dockerfiles cache (#6371) 2026-02-18 18:56:13 +02:00
Janelle Cai ef2583a9da
feat: add prefix groups to rust mooncake bench (#6308)
Signed-off-by: Janelle Cai <jcai18@mit.edu>
Signed-off-by: PeaBrane <yanrpei@gmail.com>
Co-authored-by: Yan Ru Pei <yanrpei@gmail.com>
2026-02-17 23:31:07 +00:00
Tushar Sharma 04682cf892
fix: pin pydantic-settings<2.13.0 for trtllm compatibility (#6339)
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-17 13:57:08 -08:00
Dmitry Tokarev 9e8971812a
fix: Fix pre-merge (#6341)
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
2026-02-17 15:43:22 -05:00
Dmitry Tokarev ea02149e48
fix: Fixed vllm runtime image tags. Improved docker image tag selection logic (#6290)
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
Co-authored-by: Dillon Cullinan <dcullinan@nvidia.com>
2026-02-17 13:55:49 -05:00
Ran Rubin d5af6040ac
chore: container build maintenance (#6316) 2026-02-15 23:06:45 +02:00
Hyunjae Woo 233a1e9ad0
feat: Enable ModelExpress P2P weight transfer in Dynamo vLLM worker (#6186)
Signed-off-by: Hyunjae Woo <hwoo@nvidia.com>
2026-02-13 23:24:13 +00:00
Ran Rubin a582f4df07
chore: improve Docker layer caching in sglang runtime Dockerfile (#6275) 2026-02-13 17:49:49 +02:00
Graham King bbe82f182a
chore: Remove dynamo-run and mistral-rs engine (#6203)
Signed-off-by: Graham King <grahamk@nvidia.com>
2026-02-12 21:50:58 +00:00
Dmitry Tokarev 2c747d641e
fix: TRTLLM runtime tag to be 25.12-cuda13.1-runtime-ubuntu24.04 (#6247)
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
2026-02-12 15:47:03 -05:00
Dmitry Tokarev 2132fd81d4
chore: Bumped Dynamo base image tag to 25.06-cuda12.9-devel-ubuntu24.04 (#6239)
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
2026-02-12 13:38:33 -05:00
Alec 2c6e6d2204
chore(deps): bump vLLM to 0.15.1 (#6102)
Signed-off-by: alec-flowers <aflowers@nvidia.com>
Signed-off-by: Alec <35311602+alec-flowers@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-11 21:19:31 +00:00
Dillon Cullinan 3188c70a3b
chore: Templating Feedback Followup (#6125)
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
2026-02-11 13:09:27 -05:00
Graham King c8ad4aa673
fix: add python requirements for vllm tests (#6128)
Signed-off-by: Graham King <grahamk@nvidia.com>
2026-02-10 23:10:18 +00:00
milesial bf9e6d04aa
chore: remove media-nixl feature (#5940)
Signed-off-by: Alexandre Milesi <milesial@users.noreply.github.com>
2026-02-10 22:58:30 +00:00
Dillon Cullinan ac0206293c
feat: Dockerfile templating (#5633)
Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>
2026-02-10 08:56:17 -05:00
KrishnanPrash e7936c2516
fix(sglang): remove apt-installed python3-blinker (#5995)
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com>
2026-02-06 17:07:40 -06:00
Tushar Sharma 6401e34daf
ci: Transition deploy tests to pytest framework (#5874)
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
2026-02-06 21:09:34 +00:00
Anant Sharma 20847a6e4e
ci: xfail failing chunked_prefill_offload test and fix ucx tag (#6006)
Signed-off-by: Anant Sharma <anants@nvidia.com>
2026-02-05 13:08:32 -05:00
Ran Rubin 6ac17b995a
ci: Using Dynamo Builder (#5914) 2026-02-05 11:22:56 +02:00
Yan Ru Pei 50e17783d9
fix: ignore benchmarks pytest for now (#5983)
Signed-off-by: PeaBrane <yanrpei@gmail.com>
2026-02-05 08:26:25 +00:00
Yan Ru Pei 16956ac60b
fix: update aiperf version (#5982)
Signed-off-by: PeaBrane <yanrpei@gmail.com>
2026-02-04 16:43:27 -08:00
Nate Mailhot 04ec58c7ee feat: sccache metrics (#5177)
Signed-off-by: Nate Mailhot <nmailhot@nvidia.com>
2026-02-04 09:14:27 -08:00
Keiven C f70dd6638b
fix: uv network timeout to be more resilient to intermittent network issues (part 2) (#5530)
Signed-off-by: Keiven Chang <keivenchang@users.noreply.github.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
2026-02-03 09:06:42 -08:00
Tanmay Verma ba711cc1ac
chore: Upgrade to Tensorrt-LLM 1.3.0rc1 (#5700)
Co-authored-by: Pavithra Vijayakrishnan <160681768+pvijayakrish@users.noreply.github.com>
2026-01-29 15:20:58 -08:00
Tushar Sharma c5e30afb0d
build: avoid installing development tools from yum and install specific packages from dnf (#5650)
Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
2026-01-29 00:35:29 +00:00
Alec 63bee8b303
chore: bump vLLM to 0.14.1 (#5691)
Signed-off-by: alec-flowers <aflowers@nvidia.com>
2026-01-27 19:51:29 -08:00
Kyle McGill f9858193a0
fix: Update the vllm docker image to use the cuda sampler rather than the pytorch one (#5613) 2026-01-27 04:54:22 +00:00
Keiven C 8d9ccdfb8d
fix: remove cupy-cuda13x from dev container (#5669)
Signed-off-by: Keiven Chang <keivenc@nvidia.com>
Co-authored-by: Keiven Chang <keivenchang@users.noreply.github.com>
2026-01-26 18:03:48 -08:00
Dmitry Tokarev 546f1bbc0d
fix: sglang version in build.sh and docs (#5665)
Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
2026-01-26 19:53:55 -05:00
ishandhanani 4f981e282a
feat: sglang update to 0.5.8 (#5655) 2026-01-26 23:46:05 +00:00
jh-nv 65a37708e1
chore: do not default to privileged mode when mounting workspace. (#5644) 2026-01-26 18:22:12 -05:00
Jason Zhou aecf005136
chore: use aic release/0.6.0 (#5600) 2026-01-26 13:34:10 -08:00
Alec 50f1e0e17a
chore(deps): bump vLLM to 0.14.0 (#5593)
Signed-off-by: alec-flowers <aflowers@nvidia.com>
Signed-off-by: Alec Flowers <aflowers@nvidia.com>
Signed-off-by: Alec <35311602+alec-flowers@users.noreply.github.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-24 01:19:06 +00:00