Repository navigation
[2/3] server: expose request sampling controls for Drama CUDA - #6
Draft
krunkosaurus wants to merge 2 commits into
Draft
krunkosaurus wants to merge 2 commits into
krunkosaurus wants to merge 2 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
This was referenced Oct 25, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 2 of the legacy CUDA stack; merge protected chunking #5 first. Incremental source diff.
Expose the same request controls as modern TensorFold #1277: temperature, top_p, top_k in 1..64 and exact u64 seed. Whole/streamed speech, references, CFG, long-form history and redraws retain the selected values. Each long-form segment starts with the supplied seed; retries within that segment add their attempt ordinal with wrap. Omitted/null settings preserve the existing prompt-hash/default policy.
The public MLX API is a behavior reference. Native code is independently implemented; legacy sampling arithmetic/RAS is unchanged.
Receipt
Zig 0.17.0, macOS arm64, synthetic CPU code/PCM fixtures:
Draft: no CUDA device/model, acoustic listening, audio identity or performance qualification. The base's O.DIRECT use prevents CUDA host compilation on this Mac. No guard was weakened and no speed/quality or cross-backend byte-parity claim is made.
Pinned-engine packaging and user documentation: drama-tensorfold companion PR.