Update to llama.cpp 2026-01-01 #2108
Open
+164
−259
Conversation
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
- Update llama.cpp submodule (2025-08-14 → 2026-01-01) - Remove deprecated KV cache functions (use llama_memory_* instead) - Remove llama_sampler_init_softmax (deprecated) - Add LLAMA_ROPE_TYPE_IMROPE constant - Add llama_flash_attn_type enum (AUTO/DISABLED/ENABLED) - Add llama_params_fit_status enum - Add llama_model_meta_key enum for sampling metadata - Add llama_model_params fields: no_host, no_alloc - Replace llama_context_params.flash_attn bool with flash_attn_type enum - Add 15 new API functions: - llama_max_tensor_buft_overrides - llama_n_ctx_seq - llama_model_n_embd_inp - llama_model_is_hybrid - llama_flash_attn_type_name - llama_model_meta_key_str - llama_adapter_meta_* functions (5) - llama_log_get/set - llama_memory_breakdown_print - Add ggml_log_callback typedef - Disable LLAVA build (CMake incompatibility in upstream mtmd) - Bump version 0.3.16 → 0.4.0 Breaking changes: - flash_attn bool removed, use flash_attn_type enum - KV cache functions removed, use llama_memory_* API Tested with Nemotron-3-Nano-30B hybrid model.
avion23
force-pushed
the
update-llama-cpp-2026-01
branch
from
502532a to
23c10e8
Compare
Author
|
Tested on macos using CMAKE_ARGS="-DGGML_METAL=on" pip3.14 install --force-reinstall --no-cache-dir "llama-cpp-python @ git+https://github.com/avion23/llama-cpp-python.git@update-llama-cpp-2026-01" --break-system-packages |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
Bindings were 5 months outdated, preventing newer model architectures from loading.
Updates bindings to llama.cpp commit be47fb92 (2026-01-01).
Removed
llama_kv_self_*functions (usellama_memory_*API)llama_sampler_init_softmax()Added
Enums:
LLAMA_ROPE_TYPE_IMROPEllama_flash_attn_typellama_params_fit_statusllama_model_meta_keyStruct fields:
llama_model_params:no_host,no_allocllama_context_params:flash_attn_type(replacedflash_attnbool)Functions:
llama_max_tensor_buft_overrides,llama_n_ctx_seq,llama_model_n_embd_inp,llama_model_is_hybrid,llama_flash_attn_type_name,llama_model_meta_key_str,llama_adapter_meta_*(5 functions),llama_log_get,llama_log_set,llama_memory_breakdown_printBreaking Changes
flash_attn parameter:
KV cache API:
Other
ggml_log_callbacktypedefLLAMA_INSTALL_VERSIONbefore subdirectory include)Tested: macOS ARM64 Metal, Python 3.14, Nemotron-3-Nano-30B