Upgrade llama.cpp from b9284 to b9297#191
Merged
Merged
Conversation
Patch is mostly internal (ggml backend fixes, NVFP4 MTP scale tensors, SYCL MoE counting-sort speedup, Adreno MoE boundary-check fix). The one user-visible behavioural change is in server-task.cpp: common_chat_parser_params::is_continuation (new) now gates the empty-prefill chat_msg init — non-continuation requests correctly emit the first delta instead of suppressing it. Java already wires continue_final_message through to the request JSON, so behaviour is picked up automatically. Build + all 435 C++ tests pass.
bernardladenthin
had a problem deploying
to
startgate
GitHub Actions
Error
|
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Changes
This upgrade includes several upstream improvements across multiple backends and components:
Java-relevant changes:
InferenceParameters.setContinueFinalMessage()now correctly emits the first delta for non-continuation requests instead of suppressing it (previously only worked correctly for continuation requests)ggml_backend_tensor_get_2d_async: corrects a typo in the fast-path condition that was preventing proper fallback for multi-copy getsBackend improvements (no Java changes required):
Build system:
LLAMA_BUILD_APPdefault reverted to${LLAMA_STANDALONE}(OFF for FetchContent); project's defensive pin remains in placeTest plan
Related issues / PRs
Follows the llama.cpp upgrade procedure documented in CLAUDE.md.
Checklist
CONTRIBUTING.mdandCODE_OF_CONDUCT.mdhttps://claude.ai/code/session_016Wh3BGBeMygcFUgQvcnuG6