Revīziju vēsture

Autors SHA1 Ziņojums Datums
  Johannes Gäßler f64d44a9b9 CUDA: Fixed OpenLLaMA 3b mmq, reduced compile time (#2590) 2 gadi atpakaļ
  byte-6174 b19edd54d5 Adding support for llama2.c models (#2559) 2 gadi atpakaļ
  Equim 53dc399472 server: fixed wrong variable name in timing json (#2579) 2 gadi atpakaļ
  DannyDaemonic 9ca4abed89 Handle `ENABLE_VIRTUAL_TERMINAL_PROCESSING` more gracefully on earlier versions of Windows. 2 gadi atpakaļ
  Christian Demsar e59fcb2bc1 Add --n-predict -2 for stopping generation on full context (#2565) 2 gadi atpakaļ
  Martin Krasser 1638757767 Fix grammar-based sampling issue in server (#2566) 2 gadi atpakaļ
  Sam Spilsbury 916a9acdd0 ggml-alloc: Don't try to re-use buffers of external tensors (#2562) 2 gadi atpakaļ
  grahameth ea04a4ca19 add log_callback to llama_context_params for custom logging. (#2234) 2 gadi atpakaļ
  Johannes Gäßler 25d43e0eb5 CUDA: tuned mul_mat_q kernels (#2546) 2 gadi atpakaļ
  Martin Krasser f5bfea0580 Allow passing grammar to completion endpoint (#2532) 2 gadi atpakaļ
  Johannes Gäßler acfc5478ff CUDA: tighter VRAM scratch size for 65b/70b (#2551) 2 gadi atpakaļ
  chaihahaha 7ed8d1fe7f llm.vim : multiline autocompletion, get rid of "^@" (#2543) 2 gadi atpakaļ
  Georgi Gerganov e7f94d6fdc vim : bring back simple llm.vim example 2 gadi atpakaļ
  AustinMroz 2d7baaf50f vim : streaming and more (#2495) 2 gadi atpakaļ
  klosax f3c3b4b167 Add --rope-scale parameter (#2544) 2 gadi atpakaļ
  Georgi Gerganov 93356bdb7a ggml : mul mat tweaks (#2372) 2 gadi atpakaļ
  Georgi Gerganov 60baff7c85 ggml : pad result of ggml_nbytes() 2 gadi atpakaļ
  Georgi Gerganov 9082b5dfbf ggml : change params pointer (style change) (#2539) 2 gadi atpakaļ
  Georgi Gerganov 99d29c0094 ggml : sync (custom ops) (#2537) 2 gadi atpakaļ
  Johannes Gäßler 3d9a551816 Fixed mmap prefetch for GPU offloading (#2529) 2 gadi atpakaļ
  Georgi Gerganov f6f9896ac3 metal : fix out-of-bounds access + inc concurrency nodes (#2416) 2 gadi atpakaļ
  GiviMAD 34a14b28ff [Makefile] Move ARM CFLAGS before compilation (#2536) 2 gadi atpakaļ
  Henri Vasserman 7297128db8 [Zig] Rewrite build for Zig 0.11 (#2514) 2 gadi atpakaļ
  DannyDaemonic 86c3219895 console : fix issue related to Windows 11 PowerShell console mode persistence (#2521) 2 gadi atpakaļ
  Keiichi Tabata 2e8265ae17 convert.py : add missing abstract methods for quantized data (#2491) 2 gadi atpakaļ
  Johannes Gäßler f514d1b306 CUDA: faster k-quant mul_mat_q kernels (#2525) 2 gadi atpakaļ
  Jonas Wunderlich 332311234a fix firefox autoscroll (#2519) 2 gadi atpakaļ
  Cebtenzzre 182af739c4 server: regenerate completion.js.hpp (#2515) 2 gadi atpakaļ
  Cebtenzzre 4329d1acb0 CUDA: use min compute capability of GPUs actually used (#2506) 2 gadi atpakaļ
  Cebtenzzre 02f9d96a86 CUDA: check if event is NULL before cudaStreamWaitEvent (#2505) 2 gadi atpakaļ