Commit History

Author SHA1 Message Date
  Xuan-Son Nguyen bf7b0c9725 mtmd: pad mask for qwen2.5vl (#16954) 2 months ago
  Jinyang He fcfce040e8 ggml : LoongArch fixes (#16958) 2 months ago
  Olivier Chafik ee3a5a10ad sync: minja (glm 4.6 & minmax m2 templates) (#16949) 2 months ago
  shani-f 7e994168b1 SYCL: optimized repeat_back kernel (3× fewer asm instructions, 2× faster)Feature/sycl repeat back opt (#16869) 2 months ago
  Sascha Rogmann bcfa87622a feat(webui): improve LaTeX rendering with currency detection (#16508) 2 months ago
  Shagun Bera a2054e3a8f test-backend-ops : fix segfault in moe-expert-reduce test in support mode and coverage (#16936) 2 months ago
  Sigbjørn Skjæret dd52868050 ci : disable failing riscv cross build (#16952) 2 months ago
  Zhiyong Wang 6b9a52422b model: add Janus Pro for image understanding (#16906) 2 months ago
  Georgi Gerganov 2f966b8ed8 clip : use FA (#16837) 2 months ago
  Georgi Gerganov cd5e3b5754 server : support unified cache across slots (#16736) 2 months ago
  Aldehir Rojas 87c9efc3b2 common : move gpt-oss reasoning processing to init params (#16937) 2 months ago
  Adrian Lundberg 76af40aaaa docs: remove llama_sampler_accept reference in sampling sample usage (#16920) 2 months ago
  mnehete32 7db35a7958 CUDA: add FLOOR, CEIL, ROUND, TRUNC unary ops (#16917) 2 months ago
  Aaron Teo a864132ba5 devops: fix failing s390x docker build (#16918) 2 months ago
  Aaron Teo d38d9f0877 ggml: add s390x cpu-feats (#16774) 2 months ago
  Georgi Gerganov 7fd205a8e8 scripts : add script to bench models (#16894) 2 months ago
  Pascal 2f68ce7cfd webui: auto-refresh /props on inference start to resync model metadata (#16784) 2 months ago
  Pascal e4a71599e5 webui: add HTML/JS preview support to MarkdownContent with sandboxed iframe (#16757) 2 months ago
  Adrien Gallouët dd5e8cab51 vendor : update cpp-httplib to 0.27.0 (#16846) 2 months ago
  Xuan-Son Nguyen cf659bbb8e mtmd: refactor preprocessing + support max/min pixels (#16878) 2 months ago
  Aleksander Grygier d8b860a219 Add a setting to display message generation statistics (#16901) 2 months ago
  Jaromír Hradílek 1ae74882f8 webui: recognize AsciiDoc files as valid text files (#16850) 2 months ago
  Sigbjørn Skjæret 961660b8c3 common : allow --system-prompt-file for diffusion-cli (#16903) 2 months ago
  Sigbjørn Skjæret 74fef4129f codeowners : update after refactor (#16905) 2 months ago
  Jeff Bolz 5d8bb900bc vulkan: Fix multi_add invalid descriptor usage (#16899) 2 months ago
  Jeff Bolz 2e76e01360 vulkan: fuse mul_mat+add and mul_mat_id+add_id (#16868) 2 months ago
  Oliver Simons d3dc9dd898 CUDA: Remove unneded bias/gate dims in fused mmvq (#16858) 2 months ago
  Piotr Wilkin (ilintar) bea04522ff refactor : llama-model.cpp (#16252) 2 months ago
  Piotr Wilkin (ilintar) 0de0a01576 model : Minimax M2 (#16831) 2 months ago
  Giuseppe Scrivano e58d585604 model : add Granite Hybrid nano types (#16896) 2 months ago