[MTaraszewski]

What does this PR do?

Prepare the package versions, dependency pins, changelogs, and execution-engine version metadata for Inference 1.8.2.

• Set roboflow-workflows to 0.2.13rc1, update its release pin, and regenerate its lockfile.
• Set streamvision to 0.1.3rc1 (aligned with Prepare Streamvision 0.1.3rc1 release candidate #3262), update its release pin, require roboflow-workflows>=0.2.13rc1 in its Workflows extra, and regenerate its lockfile against the published RC.
• Use the published inference-models==0.39.5 in both regenerated lockfiles and raise the Workflows minimum to that stable version. The Models package version itself is unchanged.
• Raise the Workflows SDK minimum to inference-sdk>=1.8.1, matching Bump Workflows to 0.2.13rc1 and update Models and SDK dependencies #3261; both package lockfiles select the published SDK 1.8.1.
• Set the bundled execution-engine version to 1.17.0; update the corresponding test assertions and documentation example.
• Move pending changelog entries under the new package versions and retain fresh Unreleased sections.

inference/core/version.py is already 1.8.2 on main, so it does not change in this diff.

Related Issue(s): N/A — release preparation.

Type of Change

• Other: Release preparation — package versions, dependency pins, regenerated lockfiles, changelogs, and execution-engine version metadata.

Testing

• I have tested this change locally — package-pin tests, lockfile validation, and code-quality checks.
• I have updated existing tests — execution-engine version assertions now expect 1.17.0.
• N/A — new runtime tests; this PR introduces no new runtime behavior.

Test details:

• Package-version and requirements-pin unit tests: 4 passed.
• Workflows and Streamvision lockfiles regenerated using uv lock; both passed uv lock --check --offline.
• make check_code_quality passed. Existing advisory lint warnings remain.
• git diff --check passed.
• Full integration and hosted-platform suites were not run; updating their version assertions does not establish that those suites pass.

Checklist

• My code follows the style guidelines of this project — code-quality checks passed.
• I have performed a self-review of my own code — reviewed the release metadata, dependency pins, lockfile changes, and changelog entries.
• N/A — additional code comments; the changes do not introduce runtime logic.
• My changes generate no new errors in the checks run — existing advisory warnings remain.
• I have updated the documentation — package changelogs and the execution-engine version example.

The Workflows version and dependency minimums are aligned with #3261: 0.2.13rc1, Models >=0.39.5, and SDK >=1.8.1. Dependency resolution does not require SDK 1.8.2 publication.

Remaining release work

• Validate the release candidates before publishing the final release.

The Streamvision lockfile resolves the published roboflow-workflows==0.2.13rc1, matching the top-level Inference requirements pin.

[github-actions]

👋 Thanks for the pull request! Here is how automated Claude review works here, so you spend credits (and reviewer time) wisely.

🚧 Right now this is a draft, so automated Claude review is paused — nothing is being spent yet. Mark it Ready for review to trigger it.

Warning

💸 The Claude reviewer bills in credits, not vibes

Automated review spins up a real agent that reads real code and spends real credits on every pass. It is glad to help — but it is not a rubber duck, a linter you poke in a loop, or a substitute for reading the contributing guide. Treat it like an expensive senior reviewer whose time you booked, and show up prepared.

Draft when unsure, Ready when you mean it:

• 🌱 Not sure the PR is in good shape yet? Keep it (or set it back) as a draft — drafts pause review, so you can push and iterate without burning credits on a moving target.
• 💪 Feel strong about the contents? Mark it Ready for review and the reviewer will take a look.

However you get there, arrive prepared:

• 🧱 Bring a SOLID, thorough PR. Point your local agent at our skills/ to tune it to our guidelines first — or, if you are one of those fabled carbon-based contributors, read them yourself. A half-baked diff costs exactly the same to review as a finished one.
• ✅ Resolve every comment before you re-request review. Re-requesting with threads still open means paying twice for the same conversation.
• 🔁 Do not use CI review as an inner loop for a local agent. The reviewer is not a step-by-step debugger — do the unfolding locally and arrive with the answer, not the search.
• 🙋 If something looks off, ask a human. One question to a maintainer is cheaper and faster than three rounds of agent re-review chasing a misread.

Reviews are not free. A draft costs nothing to review; a Ready PR is a promise that it is worth reviewing.

• Prefer to skip automated review entirely? Add the skip-claude-review label.

[github-actions]

🤖 Claude review started at commit 54aa9c27691b79cf2cc46e78bc57e6449f8539ba.

New commits are not auto-reviewed. Add the claude-review label to request a re-review — the label is consumed when the review starts, so just add it again next time.

[CLAassistant]

All committers have signed the CLA.