Pinned to Ideas

• 13 You must be logged in to vote

  💡

  GPU optimization across different cards

  performance Speed related topics hardware Hardware related

  JohannesGaessler started May 13, 2023 in Ideas

  

  

  

  

  16

Discussions

• 5 You must be logged in to vote

  💡

  I'm starting to implement LLaVA.cpp, but where?

  monatis started Jun 25, 2023 in Ideas

  

  

  2
• 3 You must be logged in to vote

  💡

  Llama.cpp product roadmap medium term?

  ianscrivener started Jun 19, 2023 in Ideas

  

  

  4
• 8 You must be logged in to vote

  💡

  Adding support for image vision

  duplicate This issue or pull request already exists

  rPman started Apr 22, 2023 in Ideas · Closed

  

  

  

  

  4
• 3 You must be logged in to vote

  💡

  LongMem: Augmenting Language Models with Long-Term Memory

  tmm1 started Jun 21, 2023 in Ideas

  

  

  0
• 2 You must be logged in to vote

  💡

  Using AI hardware accelerators on modern computers and possibly smartphones

  Dampfinchen started May 31, 2023 in Ideas

  

  

  1
• 2 You must be logged in to vote

  💡

  Support SIGHUP to abort running model for API servers

  spirilis started Jun 4, 2023 in Ideas · Closed

  

  

  

  

  4
• 3 You must be logged in to vote

  💡

  I had the idea of smoothing outliers before quantization, apparently it's a VERY BAD idea

  KerfuffleV2 started Jun 6, 2023 in Ideas

  

  

  1
• 3 You must be logged in to vote

  💡

  Adding output file flag to main

  dmeleedy started Jun 2, 2023 in Ideas

  

  

  0
• 4 You must be logged in to vote

  💡

  Batch processing multiple runs using a GPU with insufficient video memory (any small amount) but entirely within the GPU, i.e. faster than cpu

  rPman started Jun 1, 2023 in Ideas

  

  

  0
• 3 You must be logged in to vote

  💡

  a possible strategy for training large models

  rtkclouds started May 29, 2023 in Ideas

  

  

  1
• 3 You must be logged in to vote

  💡

  Interactive HTML Interface

  ivanstepanovftw started May 27, 2023 in Ideas

  

  

  0
• 4 You must be logged in to vote

  💡

  Multiplication Ops for int16

  pablogranolabar started May 26, 2023 in Ideas

  

  

  0
• 7 You must be logged in to vote

  💡

  discord server?

  nazthelizard122 started Mar 18, 2023 in Ideas

  

  

  

  

  10
• 1 You must be logged in to vote

  💡

  Curious, is BLOOM an intended target supported model?

  linuxmagic-mp started May 23, 2023 in Ideas

  

  

  2
• 1 You must be logged in to vote

  💡

  Voice chat support?

  Asory2010 started May 20, 2023 in Ideas

  

  

  2
• 2 You must be logged in to vote

  💡

  Change the double control-c thing

  dmeleedy started May 19, 2023 in Ideas

  

  

  3
• 1 You must be logged in to vote

  💡

  Potiential improvement that I don't have time to benchmark but it looks promising

  mikefaille started May 18, 2023 in Ideas

  

  

  1
• 1 You must be logged in to vote

  💡

  Command Mode.

  iplayfast started May 18, 2023 in Ideas

  

  

  0
• 1 You must be logged in to vote

  💡

  What about adding OpenLLaMA support?

  Sovenok-Hacker started May 11, 2023 in Ideas · Closed

  

  

  2
• 2 You must be logged in to vote

  💡

  Consider adding an example for a dynamic link library

  Calandiel started May 6, 2023 in Ideas

  

  

  3
• 3 You must be logged in to vote

  💡

  NVIDIA GPT 2B model 30x smaller than LLAMA using almost same data, free for commercial use !

  batmanonline started May 2, 2023 in Ideas

  

  

  1
• 2 You must be logged in to vote

  💡

  A talking head for llama.cpp?

  StuartIanNaylor started May 1, 2023 in Ideas

  

  

  0
• 2 You must be logged in to vote

  💡

  Refactor using ld.so hardware capabilities

  leuc started Apr 22, 2023 in Ideas

  

  

  

  

  3
• 5 You must be logged in to vote

  💡

  Add the option to train a model

  luussta started Apr 22, 2023 in Ideas

  

  

  0