6d ago · llama.cpp releases →
Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF…
Includes: b12192: mtmd: support MageVL-Cowork model (#31819), b12193: mtmd: move log-mel computation to cgraph (#31808), b12194: hexagon: n_hvx and n_threads cleanup (#31857), llama.cpp b12195, llama.cpp b12196, llama.cpp b12197, b12198: server : downgrade per-request proxy logs to trace (#31859)