Add GPU container with RTX 5070 passthrough #55
No reviewers
Labels
No labels
IaC
admin
agent-containers
automation
blockchain
cleanup
data
documentation
hardware
hotfix
infrastructure
maintenance
metrics
ml
monitoring
networking
resilience
security
tailscale
trading
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
midwitmoneymgmt/m3-infra!55
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "55-ml-container-gpu"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Changes
New
gpucontainer for GPU compute workloads (MoLE research, rvLLM, trading inference):incus_storage_volume.gpu_dataonm3-trade-pool, mounted at/var/lib/modelsincus_instance.gpuat10.121.25.32(workload range)gpudevicedefault+nixos-containerprofilesWhy workload range, not AI range
The GPU container is a compute tool, not an AI agent. It sits alongside dev (.30) and trade (.31) in the workload range. The .4x range is reserved for autonomous agents (claude-code, codex).
Post-merge
Container boots as bare NixOS with GPU access. Follow-up: configuration.nix with CUDA/nvidia packages, rvLLM, and a deploy job.
Closes #55
added 1 commit
74690c32- Rename ml -> gpu, move to workload range .32Compare with previous version
added 1 commit
e97260d9- Add gpu container with RTX 5070 passthrough at .32Compare with previous version
changed title from
Add ML container with RTX 5070 GPU passthroughtoAdd GPU container with RTX 5070 passthroughchanged the description
assigned to @AlexCaswen
mentioned in commit
7aae4a82d0