Prep 7.12.0

[mrobers1982]

Context

• The GraphQL routing fix for embedding metadata operations is already on develop via [PLT-4347] Route embedding metadata operations through GraphQL #2074, but the published SDK is still 7.11.0
• Customers hitting JSONDecodeError with opaque lbx_ API keys only get the fix after a versioned release (Prep → tag → Publish)
• This is the standard Prep step; same shape as Prep 7.10.0 #2069 / Prep 7.11.0 #2071

In scope

• Cut release 7.12.0 so pip install labelbox ships the embedding metadata GraphQL routing fix
• Document that change in the changelog for 7.12.0

Out of scope

• GitHub release/tag and PyPI Publish (next steps after merge)
• Any further SDK code changes
• The remaining direct ADV calls (import_vectors_from_file, get_imported_vector_count) that [PLT-4347] Route embedding metadata operations through GraphQL #2074 intentionally left on AdvClient

🤖 Generated with Claude Code

---

Note

Low Risk
Release-only version and documentation updates with no runtime code changes in this diff.

Overview
Bumps the labelbox package from 7.11.0 to 7.12.0 across pyproject.toml, labelbox.__version__, and the Sphinx docs release field so the next publish ships a new SDK version.

Adds a 7.12.0 changelog entry (2026-08-25) under Fixed, calling out that embedding metadata APIs (Client.create_embedding(), Client.get_embeddings(), Client.get_embedding_by_id(), Embedding.delete()) are routed through GraphQL for opaque lbx_ API keys (#2074). This PR does not change SDK implementation—only release metadata and release notes for code already on develop.

^{Reviewed by Cursor Bugbot for commit 9de436b. Bugbot is set up for automated code reviews on this repo. Configure here.}

[datadog-us5-labelbox]

⚠️ Warnings

⚠️ Your PR has warnings. Please review the issues below.

🚦 10 Pipeline jobs failed

Labelbox Python SDK Staging (Develop) | build (3.10, STAGING_API_KEY_ORG_CMOI3PQ7801GM070D4TPG7FMH, DA_GCP_LABELBOX_API_KEY) / integration

View in Datadog · View in GitHub Actions

3 failed tests. Assertion Error: 'Unique name or global key conflict' instead of expected 'Schema did not exist'.

Labelbox Python SDK Staging (Develop) | build (3.10, STAGING_API_KEY_ORG_CMOI3PQ7801GM070D4TPG7FMH, DA_GCP_LABELBOX_API_KEY) / unit-data

View in Datadog · View in GitHub Actions

MEA prediction import has been deprecated. Error: 9 FAILED_PRECONDITION

Labelbox Python SDK Staging (Develop) | build (3.11, STAGING_API_KEY_ORG_CMOI3PQ7801GM070D4TPG7FMH, DA_GCP_LABELBOX_API_KEY) / integration

View in Datadog · View in GitHub Actions

5 failed tests due to exceeding the maximum limit for custom embeddings set at 10 per organization, along with assertion errors related to non-existent schema IDs.

View all 10 failed jobs.

Useful? React with 👍 / 👎

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 9de436b | Docs | View more details | Give us feedback!}