Discussions

• 2 You must be logged in to vote

  🙏

  Does this support this original Meta LLaMA2?

  realhaik asked Aug 27, 2023 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  💬

  FasterTransformer and OneFlow for benchmarks

  vadimkantorov started Aug 17, 2023 in General

  

  

  0
• 1 You must be logged in to vote

  💬

  CTranslate2 benchmark

  francis2tm started Jul 26, 2023 in General

  

  

  0
• 1 You must be logged in to vote

  🙏

  Triton inference server deployment.

  cm2435 asked May 28, 2023 in Q&A · Closed · Unanswered

  

  

  1
• 1 You must be logged in to vote

  💬

  Tensor Parallelism

  kamalkraj started Apr 25, 2023 in General

  

  

  0
• 1 You must be logged in to vote

  🙏

  Why is heads and dhead hardcoded in flash attention test ?

  3outeille asked Apr 9, 2023 in Q&A · Answered

  

  

  1
• 1 You must be logged in to vote

  🙏

  Unable to achieve the performance in experimental/whisper/speedup.ipynb

  S-GH asked Mar 30, 2023 in Q&A · Closed · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  Getting Llama to run

  dnhkng asked Mar 24, 2023 in Q&A · Unanswered

  

  

  0
• 2 You must be logged in to vote

  💡

  Integration with Triton Inference Server

  WissamAntoun started Mar 7, 2023 in Ideas

  

  

  1
• 1 You must be logged in to vote

  📣

  Welcome to kernl Discussions!

  white-gorilla announced Jan 12, 2023 in Announcements

  

  

  0
• 1 You must be logged in to vote

  💬

  Will kernl support huggingface's codegen model?

  songhuan541 started Feb 27, 2023 in General

  

  

  1
• 1 You must be logged in to vote

  💬

  Will this work with huggingface's PEFT?

  question Further information is requested

  p-christ started Feb 26, 2023 in General

  

  

  1
• 1 You must be logged in to vote

  💬

  Support for int8 matrix multiplication

  hchoi-moveworks started Feb 16, 2023 in General

  

  

  1
• 3 You must be logged in to vote

  🙏

  Python 3.9 requirement

  hchoi-moveworks asked Feb 8, 2023 in Q&A · Unanswered

  

  

  4
• 2 You must be logged in to vote

  🙏

  Why does kernl not support gpus before ampere architecture?

  seastar105 asked Feb 8, 2023 in Q&A · Answered

  

  

  1
• 1 You must be logged in to vote

  🙏

  What are "Warmup" step requirements?

  Pavelrst asked Jan 24, 2023 in Q&A · Unanswered

  

  

  3
• 3 You must be logged in to vote

  🙏

  Can't reproduce example from t5 e2e

  Pavelrst asked Jan 23, 2023 in Q&A · Unanswered

  

  

  2