vLLM
/
vllm-spyre
mirror of https://github.com/vllm-project/vllm-spyre.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
Community maintained hardware plugin for vLLM on Spyre
194 Commits 30 Branches 10 Tags 3 MiB
Python 97.1%
Shell 2.9%
Go to file
Sophie du Couédic 639295d666
[CB] Update CB docs + Refactoring scheduling step-by-step inference tests (#323) 
Two main changes in this PR:

* Scheduling tests refactoring: convert previously unit-test
`test_scheduler_cb_steps_tkv` to a helper function, and the previously
parameters functions of `test_scheduler_cb_steps_tkv` to unit tests
* Build doc from tests docstring (continuous batching only)

Closes #318 
Closes #319

---------

Signed-off-by: Sophie du Couédic <sop@zurich.ibm.com>
 		2025-07-18 10:46:57 -07:00
.github 🐛 Workaround ray issue in tests (#307) 2025-07-15 16:30:08 -06:00
docker ⬆️ bump base image (#296) 2025-07-09 14:53:23 -06:00
docs [CB] Update CB docs + Refactoring scheduling step-by-step inference tests (#323) 2025-07-18 10:46:57 -07:00
examples make truncation of token lists optional in example script (#317) 2025-07-16 22:17:05 -03:00
tests [CB] Update CB docs + Refactoring scheduling step-by-step inference tests (#323) 2025-07-18 10:46:57 -07:00
tools [docs] add debugging docs (#235) 2025-06-17 17:22:59 +00:00
vllm_spyre ✅ add assertions for warmup mode context (#294) 2025-07-17 09:37:49 -07:00
.gitignore ✨ add top-level spyre version (#224) 2025-06-09 20:47:03 +00:00
.readthedocs.yaml [Docs] Migrate from Sphinx to MkDocs (#189) 2025-06-02 19:00:49 +00:00
.shellcheckrc Required changes to put vLLM spyre support in a plugin (#2) 2025-02-18 14:56:52 +01:00
.yapfignore 👷 use UV package manager (#63) 2025-04-03 09:12:15 -06:00
CODE_OF_CONDUCT.md Add OS-related docs (#152) 2025-05-14 16:14:13 -04:00
CONTRIBUTING.md Add OS-related docs (#152) 2025-05-14 16:14:13 -04:00
DCO Add OS-related docs (#152) 2025-05-14 16:14:13 -04:00
LICENSE Initial commit 2025-01-22 07:21:21 -06:00
README.md [Docs] Update main README.md (#200) 2025-06-03 08:54:52 -06:00
RELEASING.md 📝 Add release docs (#124) 2025-04-30 15:53:32 -06:00
_local_envs_for_test.sh 🍱 Swap tests to tiny granite (#264) 2025-06-27 13:26:48 -06:00
format.sh [CI] Minor cleanup and more consistent workflow names (#158) 2025-05-23 18:38:19 +00:00
mkdocs.yaml [Docs] Migrate from Sphinx to MkDocs (#189) 2025-06-02 19:00:49 +00:00
pyproject.toml 🐛 add pytest-forked dev dep back (#297) 2025-07-09 15:22:56 -06:00
uv.lock 🐛 add pytest-forked dev dep back (#297) 2025-07-09 15:22:56 -06:00

README.md

Spyre Plugin for vLLM

| Documentation | Users Forum | #sig-spyre |

IBM Spyre is the first production-grade Artificial Intelligence Unit (AIU) accelerator born out of the IBM Research AIU family, and is part of a long-term strategy of developing novel architectures and full-stack technology solutions for the emerging space of generative AI. Spyre builds on the foundation of IBM’s internal AIU research and delivers a scalable, efficient architecture for accelerating AI in enterprise environments.

The vLLM Spyre plugin (vllm-spyre) is a dedicated backend extension that enables seamless integration of IBM Spyre Accelerator with vLLM. It follows the architecture described in vLLM's Plugin System, making it easy to integrate IBM's advanced AI acceleration into existing vLLM workflows.

For more information, check out the following:

Getting Started

Visit our documentation:

Contributing

We welcome and value any contributions and collaborations. Please check out Contributing to vLLM Spyre for how to get involved.

Contact

You can reach out for discussion or support in the #sig-spyre channel in the vLLM Slack workspace or by opening an issue.