WIP: llama.cpp-v0.6.0 #5

Draft
gitdode wants to merge 3 commits from llama.cpp-v0.6.0 into main
Owner
No description provided.
- There's now a RAII wrapper for llama_batch_ext
- Lower context size, still suitable for the purpose here
- Increase batch size; seems necessary for higher resolution images
Comment setting g_params.n_ubatch = 2048
Some checks failed
C/C++ CI / build (pull_request) Failing after 9s
b70b5621e4
This doesn't seem to be necessary anymore to avoid problems with higher
resolution images. It is not because of the new "capping image_max_tokens"
since both models tested with (GLM OCR and Gemma E4B) are not non-casual
models.
Some checks failed
C/C++ CI / build (pull_request) Failing after 9s
This pull request is marked as a work in progress.
View command line instructions

Checkout

From your project repository, check out a new branch and test the changes.
git fetch -u origin llama.cpp-v0.6.0:llama.cpp-v0.6.0
git switch llama.cpp-v0.6.0

Merge

Merge the changes and update on Forgejo.
git switch main
git merge --no-ff llama.cpp-v0.6.0
git switch llama.cpp-v0.6.0
git rebase main
git switch main
git merge --ff-only llama.cpp-v0.6.0
git switch llama.cpp-v0.6.0
git rebase main
git switch main
git merge --no-ff llama.cpp-v0.6.0
git switch main
git merge --squash llama.cpp-v0.6.0
git switch main
git merge --ff-only llama.cpp-v0.6.0
git switch main
git merge llama.cpp-v0.6.0
git push origin main
Sign in to join this conversation.
No reviewers
No labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
gitdode/ocr-cpp!5
No description provided.