Skip to content

docs(model-workflows): align cache file lists and note GPU smoke pre-pull - #1440

Merged
michael-balint merged 1 commit into
masterfrom
dholt/model-workflow-doc-notes
Nov 10, 2026
Merged

michael-balint merged 1 commit into
masterfrom
dholt/model-workflow-doc-notes

Conversation

@dholt

@dholt dholt commented Nov 6, 2026

Copy link
Copy Markdown
Contributor

Follow-up docs/comment fixes from review notes on #1423 and #1439. No behavior change.

  • vllm-quickstart: "verified on " wording for the vLLM and model licence checks.
  • Cache file lists: explain that the serving example validates with three files while fine-tuning readiness also requires tokenizer_config.json (noted in vllm-quickstart, serve-model skill, shared-cache and finetune-readiness).
  • --gpu-smoke: document that the 5s scheduler window is too short for an image pull, so pre-pull the pinned image (finetune-readiness, finetune-model skill).
  • gpu_count(): comment that it fails closed if Slurm TRES output changes.

Tests: pytest scripts/validation/tests 197 passed, 7 skipped.

…pull

- vllm-quickstart: fix 'verified on' date phrasing
- Explain why serve (3 files) and fine-tune readiness (adds
  tokenizer_config.json) differ; note it in serve-model, shared-cache
  and finetune-readiness
- Document that --gpu-smoke's 5s scheduler window requires a pre-pulled image
- Comment that gpu_count() fails closed on unexpected Slurm output

Signed-off-by: Doug Holt <[email protected]>
@dholt
dholt requested a review from michael-balint November 6, 2026 13:25
@michael-balint
michael-balint merged commit 6d07f54 into master Nov 10, 2026
32 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants