Repository navigation
[3/3] server: preserve dialogue speaker tags across long-form requests - #7
Draft
krunkosaurus wants to merge 3 commits into
Draft
krunkosaurus wants to merge 3 commits into
krunkosaurus wants to merge 3 commits into
Conversation
Keep complete special tokens, bracketed cues and nested paired tags intact across sentence, clause, space and code-point splitting. Protect topic-shift preprocessing and recognize CJK boundaries without changing prompt or frame ceilings. Exercise exact whole and streamed prompt segments, oversized spans, malformed controls and redraws. Drain inherited socket fixtures before joining so CPU validation completes on macOS. Refs: plotarmordev/drama-tensorfold#25
5 of 6 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What this changes
Part 3 of the legacy CUDA stack; merge chunking #5 then sampling #6. Incremental source diff.
Accept canonical speaker tags 0..15 in requests, references, catalogs and the native writer, and carry the active role across long-form continuations, history, paragraph shifts and redraws. Original segment slices remain intact; added role text is counted at prompt admission. Existing malformed-tag, body/file/prompt and codebook limits remain enforced. The legacy formats have no new fixed reference-count cap.
The public MLX prompt tests are a behavior reference. Modern source companion #1278 extends the corresponding current route. Native changes are independently implemented.
Receipt and adversarial review
Zig 0.17.0, macOS arm64; synthetic CPU code/PCM fixtures.
9e782dc9:zig build test-server-cpu -j2 --summary allpasses 379/383, four skipped, zero failures.abcdefghijklmnop<|speaker:15|>abcdefghijklmnopqrstuvwxyzat target 20. The stack includes protected chunking first. A composed HTTP regression now asserts all three complete prompt segments, carried speaker/CFG text, sampling settings and an exact seed above 2^53, for whole and streamed replies.Draft: no CUDA device/model inference, acoustic identity/listening, audio identity or performance qualification. The existing macOS O.DIRECT compilation limitation remains. Speaker tags preserve model roles; distinct or faithfully cloned acoustic voices are not guaranteed or claimed.
Pinned-engine packaging and user documentation: drama-tensorfold companion PR.