Commit Verlauf

Autor SHA1 Nachricht Datum
  hksdpc255 1920345c3b common : Generalized XML-style tool-call parsing with streaming support (GLM 4.5/4.6 + MiniMax M2 + SeedOSS + Kimi-K2 + Qwen3-Coder + Apriel-1.5 + Xiaomi-MiMo) (#16932) vor 2 Monaten
  jiahao su 561a3e2788 ci : change the openEuler-310p image to fix release (#17361) vor 2 Monaten
  Georgi Gerganov f40a2e5f11 gitignore : be more specific about ignored stuff (#17354) vor 2 Monaten
  Chenguang Li bc4064cfea CANN: fix acl_tensor_ptr usage in ASCEND_310P ROPE (#17347) vor 2 Monaten
  o7si 97cb3fd5ae fix: resolve undefined variable 'svr' compilation error (#17348) vor 2 Monaten
  jiahao su ffa277a54c CANN: Add openEuler-cann in build and release (#17192) vor 2 Monaten
  Jeff Bolz da95bf2a85 vulkan: support noncontig i32 copy (#17328) vor 2 Monaten
  Xuan-Son Nguyen 0de8878c96 server: split HTTP into its own interface (#17216) vor 2 Monaten
  Ruben Ortlam 38e2c1b412 vulkan: add log RTE support to fix Nvidia CI (#17320) vor 2 Monaten
  Adrien Gallouët cb44fc84e8 cmake : fix ARM feature verification (#17170) vor 2 Monaten
  Adrien Gallouët cb623de3fc ggml : add missing AVX512 feature checks (#17270) vor 2 Monaten
  Georgi Gerganov 7aaeedc098 metal : support I32 -> I32 copy (#17317) vor 2 Monaten
  Georgi Gerganov 3347e6d904 metal : faster argsort (#17315) vor 2 Monaten
  Georgi Gerganov 1a139644a8 metal : add cumsum (#17305) vor 2 Monaten
  hipudding 2376b7758c CANN: Use smart pointers to manage ACL objects (#17238) vor 2 Monaten
  Pavels Zaicenkovs dbed61294a vulkan: add LOG operation support for F32 and F16 (#17183) vor 2 Monaten
  Ruben Ortlam 80deff3648 vulkan: fix MMQ quantize_y condition (#17301) vor 2 Monaten
  Eve 8b1c339bd2 ci : revert #16249 (#17303) vor 2 Monaten
  Georgi Gerganov 416e7c7f47 metal : remove obosolete asserts (#17295) vor 2 Monaten
  Georgi Gerganov 5b2093becc server : handle context overflow during decode (#17267) vor 2 Monaten
  lhez 52e5d421f1 opencl: fix rms_norm_mul (#17250) vor 2 Monaten
  shaofeiqi 4db5641210 opencl: add kernel to handle mat mul in attention to improve encoding speed (#17181) vor 2 Monaten
  shani-f 72bd7321a7 sycl : unify unary kernels with a generic implementation and enable wide operator support (#17213) vor 2 Monaten
  Aleksander Grygier 22e1ce2f81 webui: Fix clickability around chat processing statistics UI (#17278) vor 2 Monaten
  Pascal 1411d9275a webui: add OAI-Compat Harmony tool-call streaming visualization and persistence in chat UI (#16618) vor 2 Monaten
  Sigbjørn Skjæret 662192e1dc convert : remove unnecessary chat template patching (#17289) vor 2 Monaten
  Jeff Bolz 24dc769f1b vulkan: Fuse mul_mat_id+add_id+mul and mul_mat+add+add. (#17287) vor 2 Monaten
  Ruben Ortlam 4dca015b7e vulkan: Replace 16-bit unpack8 calls to work around legacy Windows AMD driver bug (#17285) vor 2 Monaten
  Sigbjørn Skjæret 9a8860cf5d convert : use all parts in safetensors index (#17286) vor 2 Monaten
  Sigbjørn Skjæret 9d3ef4809f convert : set expert gating func in base class (#17279) vor 2 Monaten