hyperonym/basaran

Issues

• 401 error on llama2 model while access granted

  #289 opened a year ago by tomtomtomtom44
  0

• Runpod Serverless

  #288 opened a year ago by stonejohnson
  1

• Loading basaran on multiple gpus leads to error

  #280 opened a year ago by tanmaylaud
  0

• Can you create a Gradio app for this Trained model

  #276 opened a year ago by yhammadmufin
  0

• Suggestion: Passthrough to OpenAI GPT3.5 for testing

  #273 opened a year ago by FetchFast
  0

• Add support for chat completion API

  #140 opened 2 years ago by peakji
  1

• Tried multiple different models but get "The model weights are not tied..." error every time..

  #266 opened a year ago by jontstaz
  0

• How to send Audio Inputs to the Basaran

  #234 opened a year ago by Tushar-ml
  3

• Strong need for multiple `models` in a single deployment

  #263 opened a year ago by KastanDay
  3

• Llama 2 models not working - how to pass auth token?

  #232 opened a year ago by arsaboo
  1

• TypeError: issubclass() arg 1 must be a class

  #253 opened a year ago by gsuuon
  1

• :latest version tag

  #138 opened 2 years ago by mariushosting
  2

• Use basaran API as Langchain LLM

  #256 opened a year ago by brightebyte
  0

• RuntimeError: expected scalar type Float but found Half

  #215 opened 2 years ago by DataDropp
  0

• TypeError: __init__() got an unexpected keyword argument 'load_in_4bit'

  #227 opened a year ago by tanshuai
  1

• Error when Running Vicuna's FastChat Model without GPU

  #223 opened a year ago by davyeu
  1

• FR support for using fine tuned models that use Peft

  #221 opened a year ago by samos123
  0

• Langchain Prompt Format

  #198 opened 2 years ago by 0xDigest
  5

• I want use the function prefix_allowed_tokens_fn, where of basaran's source code shall I modify?

  #220 opened a year ago by zoubaihan
  1

• Falcon 40B : too slow and random answers

  #204 opened 2 years ago by ArnaudHureaux
  7

• GPTQ & 4bit

  #180 opened 2 years ago by olihough86
  3

• crash when running mosaicml/mpt-7b-* models: KeyError: 'attention_mask'

  #213 opened 2 years ago by tarasglek
  2

• QLoRa support

  #202 opened 2 years ago by bitnom
  4

• concurrent request supported?

  #205 opened 2 years ago by hudengjunai
  1

• Support for `v1/embeddings` endpoint

  #179 opened 2 years ago by josephrocca
  12

• in stream mode, the English word has no space after detokenizer and Chinese were messed up

  #197 opened 2 years ago by lucasjinreal
  14

• Vicuna problem

  #160 opened 2 years ago by zhound420
  13

• Docker run runs and then exits, does not set up server

  #183 opened 2 years ago by handrew
  2

• How to set my own parameters in model.generate() in basaran?

  #193 opened 2 years ago by zoubaihan
  2

• Inference should stop if connection is aborted/closed

  #192 opened 2 years ago by josephrocca
  2

• Define chat history format using jinja template

  #141 opened 2 years ago by peakji
  2

• Do you have Discord community?

  #185 opened 2 years ago by karfly
  1

• `v1/completions` does not include `data: ` prefix when `stream:true`

  #144 opened 2 years ago by josephrocca
  5

• How can I use this with a project using OpenAI's nodejs library?

  #120 opened 2 years ago by MarkSchmidty
  2

• ValueError: Tokenizer class LLaMATokenizer does not exist or is not currently imported.

  #139 opened 2 years ago by josephrocca
  1

• How to pass `max_token_length` to `load_model` ?

  #116 opened 2 years ago by MohamedAliRashad
  1

• Support ARM Docker images

  #110 opened 2 years ago by WillBeebe
  1

• how to run model in total offline?

  #109 opened 2 years ago by gitknu
  4

• Support for llama.cpp/ggml models

  #107 opened 2 years ago by codito
  2

• RuntimeError: mat1 and mat2 shapes cannot be multiplied

  #181 opened 2 years ago by lcw99
  1

• Getting error for model when using vicuna model

  #152 opened 2 years ago by djaffer
  5

• Possible to run on M-series chips/MPS?

  #173 opened 2 years ago by fakerybakery
  3

• Question about COMPLETION_MAX_PROMPT

  #158 opened 2 years ago by nicpopovic
  5

• Any chance to support text-generation-inference as backend?

  #147 opened 2 years ago by hewr1993
  1

• The requested URL was not found on the server

  #151 opened 2 years ago by artivis
  2

• Instructions unclear

  #137 opened 2 years ago by Anonym0us33
  1

• CORS headers

  #143 opened 2 years ago by josephrocca
  7

• Replace EventSource with POST requests in playground

  #145 opened 2 years ago by fardeon
  0

• Add a chat interface to playground

  #142 opened 2 years ago by peakji
  0

• Slow Streaming

  #99 opened 2 years ago by manojpreveen
  2