Code

Few more Inference Endpoints fixes#69

Code

Merged

tengomucho merged 8 commits intomainhuggingface/optimum-tpu:mainfrom debug-tgi-ie-pt3huggingface/optimum-tpu:debug-tgi-ie-pt3Copy head branch name to clipboardJul 8, 2024

+19-9

Commits

Commits on Jul 6, 2024

• fix(TGI): correct clear request with a give batch id

  tengomucho

  committed
  3826ecc

  Copy full SHA for 3826ecc

• ci(tgi): create images when pushing on current branch

  tengomucho

  committed
  26e1193

  Copy full SHA for 26e1193

• fix(generator): raise error if prefill receives too many requests

  tengomucho

  committed
  ba591ed

  Copy full SHA for ba591ed

• feat(tgi): add more prefill lenghts

  Show description for 30184a1
  tengomucho

  committed
  30184a1

  Copy full SHA for 30184a1

• Revert "ci(tgi): create images when pushing on current branch"

  Show description for a9ce7fc
  tengomucho

  committed
  a9ce7fc

  Copy full SHA for a9ce7fc

Commits on Jul 8, 2024

• fix(test): multiple decode test require max_batch_size to be > 1

  tengomucho

  committed
  ff2a1a5

  Copy full SHA for ff2a1a5

• fix(test): expected result is different when model is compiled

  Show description for 50a4513
  tengomucho

  committed
  50a4513

  Copy full SHA for 50a4513

• chore: bump to version v0.1.2

  tengomucho

  committed
  3b07302

  Copy full SHA for 3b07302