Add a script to automatically generate expanded release notes using an LLM #3080
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Outdated suggestions cannot be applied.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.
The PR adds a script to generate release notes that we used to prepare by hand with an LLM. It takes a draft release with automatic release notes and expands it. The LLM gets all the relevant PR descriptions in the context and is able to produce good results given the PR descriptions are sufficient.
Currently adding to be used manually – then we can automatic draft generation and editing and add to the CI on release.
Example:
"""
python scripts/release_notes.py 0.19.25rc1
Service Probes
When deploying services with rolling updates, new replicas are only registered to receive traffic after all their probes pass, preventing service disruptions during deployments.
Elastic Fleets
Fleet configuration now supports elastic scaling with improved instance selection logic. The system now prioritizes fleets that can accommodate runs without requiring additional provisioning, falling back to fleets specified in configuration when no idle instances are available. This improves resource utilization and reduces startup times for runs.
GPU Offer Aggregation
The
dstack offer
command now supports aggregating GPU information for better resource planning:This groups available GPU offers by backend, region, and GPU count, making it easier to understand resource availability across your infrastructure.
Deprecations
--local
flag is now deprecated--ssh-identity
option indstack init
is deprecated and ignoreddstack init --remove
is available to remove local repo configurations from~/.dstack/config.yml
What's Changed
DCGMWrapperInterface
nil check (bis) by @un-def in [shim] FixDCGMWrapperInterface
nil check (bis) #3001dstack offer
to aggregate GPU information by @peterschmidt85 in [Feature]: Allowdstack offer
to aggregate GPU information #2992New Contributors
Full Changelog: 0.19.24...0.19.25rc1
"""