Local Ai
b8744
llama.cpp release **b8744**, published on April 10, 2026, is a build of the ggml-org/llama.cpp C/C++ LLM inference engine. Its primary change enables the reasoning budget sampler for Gemma 4 by add...
llama.cpp release b8744, published on April 10, 2026, is a build of the ggml-org/llama.cpp C/C++ LLM inference engine. Its primary change enables the reasoning budget sampler for Gemma 4 by adding thinking_start_tag and thinking_end_tag to common_chat_params_init_gemma4(), without which the reasoning budget sampler never activates for that model. The release also makes the newline after "thought" optional in the PEG parser to handle budget=0 cases, and adds a test case for empty thinking blocks (fixing issue #21487).
Related
Source: local-ai