orchestrai: run model playbooks on all devices + provision WSL for Hermes on Windows - #678
Merged
Merged
Conversation
saman-amd
force-pushed
the
saman-amd-ram128
branch
3 times, most recently
from
August 7, 2026 16:20
783f6ad to
671c620
Compare
…rmes on Windows Two related orchestrai-config changes plus a comment reflow: 1. Scope ram_128gb device-specifically so the model playbooks run on every device instead of Halo-only. lmstudio/pytorch-rocm/n8n pick a smaller model per device; vscode/gaia/openclaw run a fixed Qwen3-A3B MoE that lemonade offloads across GPU+CPU on the smaller RDNA cards. Moved from blanket extra_tags to device_extra_tags (ram_128gb on halo/halo_box only), the pattern unsloth-llms-finetuning already uses. extra_tags is now empty. 2. Provision WSL for hermes-lemonade-server on Windows: add it to extra_install_scripts (windows-wsl.ps1 + reboot before the deps phase), so app.wsl can install the Ubuntu distro that Hermes/Podman run inside. Mirrors openclaw-lemonade-server. orchestrai_trigger.py: comment-only generalization of the WSL provisioning note.
saman-amd
force-pushed
the
saman-amd-ram128
branch
from
August 7, 2026 17:00
671c620 to
10f75f5
Compare
saman-amd
changed the title
adamlam2-amd
approved these changes
Aug 7, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Two related OrchestrAI config changes (config-only), keeping the shared orchestrai files aligned.
1. Run the model playbooks on all devices (ram_128gb scoping)
A blanket
extra_tags: [ram_128gb]combined withextra_tag_devices(ram_128gb → halo/halo_box) dropped six model playbooks from every non-128GB device. But they run on all devices in the legacytest-playbooksworkflow:Moved all to
device_extra_tagstaggingram_128gbonly on halo/halo_box (the patternunsloth-llms-finetuningalready uses).extra_tagsis now empty.2. Provision WSL for Hermes on Windows
Add
hermes-lemonade-servertoextra_install_scripts(windows-wsl.ps1+ reboot before the deps phase) soapp.wslcan install the Ubuntu distro that Hermes and Podman run inside on Windows. Mirrorsopenclaw-lemonade-server. Without it, hermes Windows runs fail with "WSL is not installed".orchestrai_trigger.py: comment-only generalization of the WSL note.Config-only; no internal coordinates.