Georgi Gerganov 3637576288 server : disable speculative decoding for SWA models (#13970) преди 8 месеца
..
batched-bench b89d605a91 batched-bench : fix pp batch contents (#13492) преди 8 месеца
cvector-generator 1d36b3670b llama : move end-user examples to tools directory (#13249) преди 9 месеца
export-lora 1d36b3670b llama : move end-user examples to tools directory (#13249) преди 9 месеца
gguf-split 1d36b3670b llama : move end-user examples to tools directory (#13249) преди 9 месеца
imatrix efb8b47eda imatrix : Add --parse-special for enabling parsing of special tokens in imatrix calculation (#13389) преди 8 месеца
llama-bench 053b1539c0 threading: support for GGML_SCHED_PRIO_LOW, update thread info on Windows to avoid throttling (#12995) преди 8 месеца
main 27ebfcacba llama : do not crash if there is no CPU backend (#13395) преди 8 месеца
mtmd bfd322796c mtmd : fix memory leak in mtmd_helper_eval_chunk_single (#13961) преди 8 месеца
perplexity 51fb96b1ff context : remove logits_all flag (#13284) преди 9 месеца
quantize e5c834f718 quantize : improve tensor-type pattern matching (#13033) преди 8 месеца
rpc c508256db2 rpc : Fix build on OpenBSD (#13541) преди 8 месеца
run 53f925074d sync : vendor (#13901) преди 8 месеца
server 3637576288 server : disable speculative decoding for SWA models (#13970) преди 8 месеца
tokenize 1d36b3670b llama : move end-user examples to tools directory (#13249) преди 9 месеца
tts 53f925074d sync : vendor (#13901) преди 8 месеца
CMakeLists.txt 9b61acf060 mtmd : rename llava directory to mtmd (#13311) преди 9 месеца