LocalAI/.github
Ettore Di Giacinto 24505e57f5
feat(backends): add CUDA 13 + L4T arm64 CUDA 13 variants for vllm/vllm-omni/sglang (#9553)
* feat(backends): add CUDA 13 + L4T arm64 CUDA 13 variants for vllm/vllm-omni/sglang

Adds new build profiles mirroring the diffusers/ace-step pattern so vLLM
serving (and SGLang on arm64) can be deployed on CUDA 13 hosts and
JetPack 7 boards:

- vllm: cublas13 (PyPI cu130 channel) + l4t13 (jetson-ai-lab SBSA cu130
  prebuilt vllm + flash-attn).
- vllm-omni: cublas13 + l4t13. Floats vllm version on cu13 since vllm
  0.19+ ships cu130 wheels by default and vllm-omni tracks vllm master;
  cu12 path keeps the 0.14.0 pin to avoid disturbing existing images.
- sglang: l4t13 arm64 only — uses the prebuilt sglang wheel from the
  jetson-ai-lab SBSA cu130 index, so no source build is needed.
  Cublas13 sglang on x86_64 is intentionally deferred.

CI matrix gains five new images (-gpu-nvidia-cuda-13-vllm{,-omni},
-nvidia-l4t-cuda-13-arm64-{vllm,vllm-omni,sglang}); backend/index.yaml
gains the matching capability keys (nvidia-cuda-13, nvidia-l4t-cuda-13)
and latest/development merge entries.

Assisted-by: Claude:claude-opus-4-7 [Read] [Edit] [Write] [Bash]

* fix(backends): use unsafe-best-match index strategy on l4t13 builds

The jetson-ai-lab SBSA cu130 index lists transitive deps (decord, etc.)
at limited versions / older Python ABIs. uv defaults to the first index
that contains a package and refuses to fall through to PyPI, so sglang
l4t13 build fails resolving decord. Mirror the existing cpu sglang
profile by setting --index-strategy=unsafe-best-match on l4t13 across
the three backends, and apply it to the explicit vllm install line in
vllm-omni's install.sh (which doesn't honor EXTRA_PIP_INSTALL_FLAGS).

Assisted-by: Claude:claude-opus-4-7 [Read] [Edit] [Bash]

* fix(sglang): drop [all] extras on l4t13, floor version at 0.5.0

The [all] extra brings in outlines→decord, and decord has no aarch64
cp312 wheel on PyPI nor the jetson-ai-lab index (only legacy cp35-cp37
tags). With unsafe-best-match enabled, uv backtracked through sglang
versions trying to satisfy decord and silently landed on
sglang==0.1.16, an ancient version with an entirely different dep
tree (cloudpickle/outlines 0.0.44, etc.).

Drop [all] so decord is no longer required, and floor sglang at 0.5.0
to prevent any future resolver misfire from degrading the version
again.

Assisted-by: Claude:claude-opus-4-7 [Read] [Edit] [Bash]
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2026-04-25 12:26:29 +02:00
..
ci fix: roll out bluemonday Sanitize more widely (#3794) 2024-10-12 09:45:47 +02:00
gallery-agent fix(ci): switch gallery-agent to sigs.k8s.io/yaml (#9397) 2026-04-17 10:10:42 +02:00
ISSUE_TEMPLATE docs/examples: enhancements (#1572) 2024-01-18 19:41:08 +01:00
workflows feat(backends): add CUDA 13 + L4T arm64 CUDA 13 variants for vllm/vllm-omni/sglang (#9553) 2026-04-25 12:26:29 +02:00
bump_deps.sh feat: do not bundle llama-cpp anymore (#5790) 2025-07-18 13:24:12 +02:00
bump_docs.sh fix: github bump_docs.sh regex to drop emoji and other text (#2180) 2024-04-29 03:55:29 +00:00
check_and_update.py fix(ci): fixup checksum scanning pipeline (#3631) 2024-09-23 10:56:10 +02:00
checksum_checker.sh fix(ci): fixup correct path for check_and_update.py (#2777) 2024-07-11 23:05:43 +02:00
dependabot.yml feat: Add backend gallery (#5607) 2025-06-15 14:56:52 +02:00
FUNDING.yml Create FUNDING.yml (#725) 2023-07-09 13:39:00 +02:00
labeler.yml chore(ci): update labels 2025-02-13 09:58:19 +01:00
PULL_REQUEST_TEMPLATE.md feat(vllm): Allow to set quantization (#1094) 2023-09-22 15:52:38 +02:00
release.yml feat(p2p): Federation and AI swarms (#2723) 2024-07-08 22:04:06 +02:00
stale.yml feat: add PR template and stale configuration (#316) 2023-05-20 09:10:20 +02:00