BluzenRiver
Job scheduling for one cluster and four GPUs
A scheduler sized for small labs: queue a run, get it placed, see why it did not start when it did not start.
- Category
- Infrastructure
- Status
- Beta
- Language
- Rust
- Licence
- Apache-2.0
- Stars
- 470
- Since
- 2026
Kubernetes is a wonderful answer to a question small labs are not asking. BluzenRiver answers the smaller one: which of these four GPUs should this run go on, and why is my job still queued.
Status
Beta. Runs the lab's own hardware.
Writing
Scheduling for four GPUs
Kubernetes answers a question small labs are not asking. Notes on writing a scheduler whose main feature is explaining itself.
Related projects
BluzenHub
A registry for artefacts you actually own
Self-hostable model and dataset registry with content addressing, so a checkpoint on your cluster and one on ours are the same object or provably are not.
- Language
- Rust
- Stars
- 1.1k
- Licence
- Apache-2.0
- Category
- Infrastructure
Kotori
Vector search that fits in a process
An embedded vector index for the very common case where your corpus is large but not that large.
- Language
- Rust
- Stars
- 740
- Licence
- MIT
- Category
- Tools
Aurora Corpus
The dataset everything else is trained on
An 18M-image illustration corpus with tags, captions and quality bands, plus the whole filtering pipeline that produced it.
- Language
- Python
- Stars
- 980
- Licence
- CC-BY-4.0
- Category
- Data