Update llama.cpp submodule to 3e4bb29

[Copilot]

Updates llama.cpp submodule from f47edb8 to 3e4bb29 (~99 upstream commits).

Changes

• Submodule: f47edb8c19199bf0ab471f80e6a4783f0b43ef81 → 3e4bb2966685facd549ac99bde1e02633e024920
• BUILD.mk: Added new source files from upstream:
  • llama.cpp/src/models/exaone-moe.cpp (EXAONE MoE architecture)
  • llama.cpp/tools/mtmd/models/mobilenetv5.cpp (MobileNetV5 vision model)

Patches

All 12 patches apply cleanly with minor line offsets (1-2 lines). No patch updates required.

API Changes

Reviewed upstream changes to include/llama.h and common/common.h. llamafile code does not use modified APIs—no integration updates needed.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Update llama.cpp submodule to 3e4bb29</issue_title>
<issue_description>A new version of llama.cpp is available. Please update the submodule from f47edb8 to 3e4bb29 and make necessary adjustments.

Step 1: Update the submodule

cd llama.cpp
git fetch origin master
git checkout origin/master
cd ..
git add llama.cpp

Step 2: Verify and update patches

• Review the patches in llama.cpp.patches/patches/
• Check if each patch still applies cleanly to the new llama.cpp version
• Update any patches that have conflicts due to upstream changes
• Run the patch application script to verify:

  cd llama.cpp.patches && ./apply-patches.sh

Step 3: Update BUILD.mk dependencies

• Review llama.cpp/BUILD.mk for any new source files or dependencies added upstream
• Remove references to any deleted source files
• Ensure all new dependencies are properly included
• Check the upstream changes for new/removed files in llama.cpp/src/, llama.cpp/common/, llama.cpp/ggml/

Step 4: Update llamafile integration code

• Check if the llamafile code that calls llama.cpp server/main needs updates
• Review llamafile/ for any API changes in llama.cpp that need to be reflected
• Pay attention to changes in llama.cpp/include/ for API modifications

Verification

After making changes, verify the build works:

make clean
make -j$(nproc)

Reference

• Upstream changes: ggml-org/llama.cpp@f47edb8...3e4bb29
• Example PR with similar updates: Update llama.cpp to f47edb8 #847

---

This issue was automatically created by the update-llama-cpp workflow.</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Update llama.cpp submodule to 3e4bb29 #854

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.