elixir-nx/bumblebee

Issues

• Bumblebee.apply_tokenizer fails for empty text

  #373 opened 20 days ago by yujonglee
  2

• Using bumblebee from git is broken

  #372 opened a month ago by dennym
  2

• Error loading gpt3.5 tokenizer

  #371 opened a month ago by yujonglee
  4

• llama3 requires 2 eos_token_id's

  #367 opened 2 months ago by jkbbwr
  1

• Support Mixtral

  #365 opened 2 months ago by WebCloud
  4

• Just download a model from HuggingFace?

  #363 opened 3 months ago by lawik
  5

• Stream audio chunk by chunk to Whisper

  #261 opened 3 months ago by mat-hek
  22

• Add support for "google/gemma-7b-it"

  #357 opened 3 months ago by kurtome
  1

• CUDA 12.2 support

  #351 opened 3 months ago by lambdaofgod
  3

• Support LLaVA

  #273 opened 7 months ago by briankariuki
  10

• Huggingface repository not found - `bert-base-google-bert`

  #355 opened 3 months ago by shishirpy
  1

• Llama 2 derivative model errors expecting top_k to be provided

  #349 opened 3 months ago by brainlid
  2

• Unknown error serving Llama 2 derivative model

  #348 opened 4 months ago by brainlid
  2

• Bumblebee error when serving Mistral LLM with latest 0.5.0 release

  #347 opened 4 months ago by brainlid
  1

• Download sharded parameters in parallel

  #258 opened 8 months ago by jonatanklosko
  0

• Running Whisper using bf16 fails

  #345 opened 4 months ago by costaraphael
  1

• Parameter persistence with sharding support

  #338 opened 4 months ago by jonatanklosko
  4

• Token Classification error in Livebook

  #343 opened 4 months ago by jkwchui
  1

• [Documentation] Image struct is not correct in Livebook Examples

  #342 opened 4 months ago by andersbjorkland
  2

• Tied word embeddings

  #339 opened 4 months ago by jonatanklosko
  2

• Add annotations to QKV layers

  #291 opened 7 months ago by seanmor5
  0

• Confidences/probabilities for Whisper results

  #335 opened 4 months ago by zacharygraber
  0

• Support returning token count information

  #287 opened 4 months ago by brainlid
  9

• Error when using TinyLlama

  #325 opened 4 months ago by trickster
  2

• Featurizer different from python?

  #323 opened 5 months ago by sonic182
  10

• Weird behaviour with progress status.

  #324 opened 5 months ago by hickscorp
  2

• Model "Salesforce - Blip large" fails

  #269 opened 7 months ago by ndrean
  7

• `TextEmbedding` crashes when both Mean Pooling and `compile` opts is specified

  #315 opened 6 months ago by thiagopromano
  2

• Cannot use `whisper-*.en` models in bumblebee

  #312 opened 6 months ago by John-Goff
  3

• Improve the error message when Hugging Face resource isn't found a the referenced location.

  #313 opened 6 months ago by meanderingstream
  1

• Token Truncation differs from Transformers implementation

  #306 opened 6 months ago by thiagopromano
  0

• Finetuning Whisper with Bumblebee / Axon

  #265 opened 6 months ago by omarelb
  2

• Use tokenizer_config.json as the primary source of metadata

  #289 opened 6 months ago by jonatanklosko
  0

• Support non-deterministic output in text generation serving

  #284 opened 6 months ago by jonatanklosko
  0

• Allow text completion streaming true/false option on a per-call basis

  #295 opened 6 months ago by brainlid
  2

• Text completion behavior is different when streaming vs not streaming

  #296 opened 6 months ago by brainlid
  1

• Support Starcoder Model

  #279 opened 6 months ago by jonastemplestein
  1

• Bumblebee fails to load gpt models in main.

  #293 opened 7 months ago by ityonemo
  1

• running 'elixir speech_to_text.exs' UndefinedFunctionError

  #292 opened 7 months ago by bartonhammond
  5

• Halting Nx Serving streams with a stop token

  #288 opened 7 months ago by zblanco
  4

• Support temperature in generation options

  #286 opened 7 months ago by jonatanklosko
  0

• Support DeepSeek Coder Model

  #278 opened 7 months ago by jonastemplestein
  11

• Multimodal projected embeddings

  #274 opened 7 months ago by lambadalambda
  2

• "cannot perform operation across devices mps and cpu" when running examples with torchx and mps

  #275 opened 7 months ago by lambadalambda
  1

• Way to tell if Bumblebee or Nx or whatever has access to a GPU?

  #272 opened 7 months ago by brainlid
  8

• How to load .en variants under recent versions?

  #267 opened 7 months ago by lawik
  4

• How do you deploy this on Fly.io?

  #270 opened 7 months ago by ndrean
  5

• Error running Stable Diffusion example

  #266 opened 8 months ago by gordoneliel
  7

• Will there be any support for LLaMa2 in the future?

  #260 opened 8 months ago by sutgeorge
  2

• Reducing entry barrier for Elixir ML - can we showcase bumblebee based elixir code exposed as REST api to integrate with other frameworks ?

  #262 opened 8 months ago by insidesmart
  2