Skip to content

Update llama.swift to llama.cpp 1.0.0 - #428

Merged
mattt merged 3 commits into
mainfrom
mattt/update-llama-swift
Nov 18, 2026
Merged

mattt merged 3 commits into
mainfrom
mattt/update-llama-swift

Conversation

@mattt

@mattt mattt commented Nov 18, 2026 •

Copy link
Copy Markdown
Collaborator

This PR moves the Llama trait to llama.swift 3.0.0, which wraps llama.cpp v1.0.0 (nightly b12359). The release removed the deprecated C API and added a trailing mtmd_helper_init_opt argument to mtmd_helper_bitmap_init_from_buf, which LlamaLanguageModel calls to decode image input. The backend now passes the default options; they only configure video decoding, and the backend rejects video input.

One behavior changes on macOS: with ffmpeg on PATH, llama.cpp now decodes WebP image data as an image, where the previous build treated it as video and the backend threw unsupportedFeature. Package.resolved doesn't change, because it has no pin for llama.swift, which only the Llama trait uses.

The Swift 6.4 CI jobs now also remove SwiftExplicitPrecompiledModules after restoring the cache: the new XCFramework's headers keep their release's timestamps, so Swift Build reused the module built from the old headers and failed to compile LlamaLanguageModel.

@mattt
mattt merged commit 1325650 into main Nov 18, 2026
14 checks passed
@mattt
mattt deleted the mattt/update-llama-swift branch November 18, 2026 14:14
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant