Conversation
Owner
Author
pd95
force-pushed
the
feature/gemma4-mlx-multi-media
branch
2 times, most recently
from
September 5, 2026 09:33
1b692a7 to
e2ce337
Compare
pd95
changed the base branch from
feature/gemma4-mlx-image-audio
to
feature/apertus-mlx
September 5, 2026 09:42
pd95
force-pushed
the
feature/gemma4-mlx-multi-media
branch
from
September 14, 2026 06:09
e2ce337 to
c17a130
Compare
pd95
removed this pull request from stack #26
September 14, 2026 18:24
pd95
added this pull request to stack #27
September 14, 2026 18:24
This was referenced Sep 14, 2026
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Codex <noreply@openai.com>
pd95
force-pushed
the
feature/gemma4-mlx-multi-media
branch
from
September 18, 2026 12:17
c17a130 to
276e227
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation
Media requests can expand substantially while being decoded or converted.
Without explicit bounds, one request can retain excessive decoded audio,
oversized WebP intermediates, or too many individually valid media items before
model execution begins. Format handling was also inconsistent between the CLI,
Gemma, MLX, and llama.cpp request paths.
These are shared request-boundary and interoperability concerns rather than a
single model's implementation details.
What this changes
miniaudiowrapper with explicitinput/output limits and a clear no-CGO rejection path;
inputs, not only a limit on each attachment;
Replacement, dependency, and scope
Official Ollama
v0.34.0contains the core Gemma 4 image/audio implementationpreviously proposed in closed PR #22. This PR supersedes #22, is based on
Apertus PR #6, and carries only the remaining shared-media work. It preserves
upstream's bounded chunking and 600-second audio cap.
Compare:
feature/apertus-mlx...feature/gemma4-mlx-multi-media
Validation
CGO-enabled media, CLI, file-data, LLM, and Gemma tests pass, as does the
explicit no-CGO MP3 rejection path. Required Linux lint/build and serialized
Darwin MLX build gates also pass. The vendored
miniaudio.hwhitespace is theonly approved diff-hygiene exception.
Packaged ordered-media runtime validation remains part of the release gate.
Refresh and AI assistance
Adapted onto Ollama
v0.34.0; headc17a130842ad9555350e8c49a621776394771a37. Codex materially assistedimplementation review, tests, refresh validation, and this PR description.