parallelConstraintDecoding

What it is

Fill a whole JSON schema of booleans and enums in two forward passes instead of generating it token by token — with calibrated confidence per field and output that is schema-valid by construction.

What the README says it does

• The disagreement is not the decoding: scoring choices by their first
• It was the prompt. The two engines originally saw different prompts, and a
• Only track is really inferable from an abstract (8 classes, majority

From the project’s README.

Where the decision sits

parallelConstraintDecoding runs its own weights or its own runtime rather than calling a hosted API, so its decisions do not depend on a vendor being up or on a key being funded. The trade is that the model is now yours to keep current.

Models it references: Qwen-2.5-1B-RLCD, Qwen-2.5-1B-RLCD.

How active it is

Created 2026-09-17, last pushed 2026-09-29, 15 stars and 1 forks when we read it on 2026-11-02. Everything in this category is weeks old at most, so a stale push date means something different here than it would on a mature project. Check the repository before you depend on it.