Skip to content

docs: record the anonymous install from the public repository #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
planetf1 merged 1 commit into from
Oct 22, 2026
Merged

docs: record the anonymous install from the public repository #1

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion docs/STATUS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@ and `ibm-granite/granite-switch-4.1-8b` unless a row names another checkpoint.
| FlashAttention 3 | FA2 and FA3 servers side by side: the 800-item comparison, 40 cached one-token reads (the case in granite-switch#139), load at 8 and 32 clients | no difference beyond batching noise in answers or speed; FA2 kept in `serve.sh` while #139 is open |
| Load | 1 to 128 concurrent clients, three-question requests | about 140 requests/s plateau, no errors |
| Plugin-only install | the README's steps in a fresh environment, then its `vllm serve` command and the e2e battery | pass |
| Install from this repository | `pip install git+https://github.com/generative-computing/granite-switch-systemone` into a fresh venv, over HTTPS and SSH with authenticated access while the repository was private; entry-point check; the unit and spec suites against the installed copy from a neutral directory | pass, 184 tests; not yet repeated anonymously, which needs the repository to be public |
| Install from this repository | `pip install git+https://github.com/generative-computing/granite-switch-systemone` into a fresh venv, over HTTPS and SSH with authenticated access while the repository was private, with the entry-point check and the unit and spec suites against the installed copy from a neutral directory (184 tests, pass); then anonymously (no git config, credentials or SSH agent) once public, at v0.1.2 | the anonymous install resolved and registered the `vllm.endpoint_plugins` entry point; the suites were not rerun on it |
| vLLM 0.26.0, 0.27.1, 0.28.0, 0.29.0 live (at `v0.1.0` only; not repeated for later changes) | one server per version on one H100 (torch 2.11.0+cu130 on 0.26, torch 2.13.0+cu130 on the rest): the e2e battery, `probe_check`, `test_decisions_e2e`, `test_prefix_sharing`, and the unit suite under each version's own venv | 11/11 e2e and every suite passes on all four versions |
| The 30b checkpoint | `granite-switch-4.1-30b` on vLLM 0.30.0, `--tensor-parallel-size 2` on two H100s: the full battery | pass |
| Tensor parallelism | the 8b with `--tensor-parallel-size 2`: the full battery | pass |
Expand Down
Loading