-
Notifications
You must be signed in to change notification settings - Fork 204
Open
Description
High priority:
- [Feature]: Default
working_dirto the image's working directory #3124 - [Meta]: Fleet-first UX #2973
- [Meta] Improve
kubernetesbackend #3126 - Sub-projects (allow share SSH fleets, backends, gateways across subprojects, and eventually quotas)
- [Feature]: Events (formerly Audit Logs) #3290
- Better cluster support
Experimental:
- Add SGLang Router Support #3267
- [Feature]: TTFT/ITL Based Autoscaling #3293
- [Feature]: shim and runner update mechanism #3288
Backlog:
- GCP DWS flex support
- Performance & Pydantic v2
- Quotas (pending on sub-projects)
Unsorted backlog:
- MIG support
- Better volumes support
- Multi-tenancy
- vLLM router integration (after SGLang)
- [Feature]: Support IAP tunneling when connecting to GCP VMs #2554 [Feature]: Allow to use SSH via AWS Session Manager #2562
nikita-toffee-ainikita-toffee-ai
Metadata
Metadata
Assignees
Labels
No labels