premAI-io/prem-services

Issues

• Mistral 7B Instruct - "cannot parse response" after one or two response

  #142 opened 10 months ago by Janaka-Steph
  5

• Mistral 7B Instruct - Tokens <|im_end|> <|im_start|> in response

  #143 opened 10 months ago by Janaka-Steph
  0

• Chat: uplift errors of model on user requests

  #144 opened 10 months ago by tiero
  0

• Mistral 128k: does not support more than 512 tokens

  #145 opened 10 months ago by tiero
  0

• Mistra Instruct weird error

  #146 opened 10 months ago by tiero
  0

• Stable Beluga generation issue

  #139 opened 10 months ago by filopedraz
  1

• drop `./ml/models`

  #140 opened 10 months ago by casperdcl
  0

• vec-redis docker

  #135 opened a year ago by casperdcl
  0

• Check CI not building binaries

  #134 opened a year ago by tiero
  0

• Enable streaming response for Petals services

  #114 opened a year ago by filopedraz
  0

• llama-2-13b-chat does not expose a /docs endpoint

  #128 opened a year ago by tiero
  1

• make internal port configurable

  #120 opened a year ago by casperdcl
  0

• Compile Petals services

  #115 opened a year ago by filopedraz
  0

• Download Petals models from Torrent

  #113 opened a year ago by filopedraz
  2

• Issue: image generation quality for stable diffusion models dropped significantly

  #89 opened a year ago by filopedraz
  0

• Serve Wizard Models

  #52 opened a year ago by filopedraz
  0

• Serve OCR Models

  #53 opened a year ago by filopedraz
  1

• Serve Upscalers

  #46 opened a year ago by filopedraz
  0

• Serve uncensored Models

  #103 opened a year ago by filopedraz
  0

• Serve Stable Audio

  #102 opened a year ago by filopedraz
  0

• Serve Xwin LM

  #109 opened a year ago by filopedraz
  0

• Vicuna and GPT4All not working on arm

  #101 opened a year ago by filopedraz
  0

• Serve Tokenflow

  #99 opened a year ago by filopedraz
  0

• Serve code llama

  #94 opened a year ago by filopedraz
  0

• Serve Codellama OpenAssistant

  #97 opened a year ago by filopedraz
  0

• Serve StableCode

  #87 opened a year ago by filopedraz
  0

• Serve IDEFICS

  #88 opened a year ago by nsosio
  0

• Serve all TabbyML Models

  #86 opened a year ago by filopedraz
  0

• Expose and serve more Image Edit models

  #84 opened a year ago by filopedraz
  0

• Serve Perfusion

  #85 opened a year ago by filopedraz
  0

• fix dfs-dalle

  #66 opened a year ago by casperdcl
  0

• fix cht-dolly-v2

  #67 opened a year ago by casperdcl
  0

• Serve Qwen 7B

  #83 opened a year ago by filopedraz
  0

• Serve Orca

  #64 opened a year ago by filopedraz
  0

• Serve h20 Models

  #65 opened a year ago by filopedraz
  0

• Serve most famous ggml models

  #58 opened a year ago by filopedraz
  1

• Serve Free Willy Models

  #62 opened a year ago by filopedraz
  0

• Serve Llama 2 70b

  #63 opened a year ago by filopedraz
  0

• Serve llama 2

  #56 opened a year ago by filopedraz
  2

• CD: fix ghcr push error

  #81 opened a year ago by casperdcl
  2

• Support multiple GPUs

  #98 opened a year ago by filopedraz
  0

• fix edb-all-minilm

  #91 opened a year ago by casperdcl
  1

• CD: investigate `buildx` errors

  #80 opened a year ago by casperdcl
  1

• Enable stream properly for HF based models

  #54 opened a year ago by filopedraz
  0

• remove GPU toolkit from CPU images

  #82 opened a year ago by casperdcl
  0

• `generations` endpoint in chat services is not using past messages as prompt in successive turns

  #69 opened a year ago by biswaroop1547
  0

• CI: auto-manage runners

  #70 opened a year ago by casperdcl
  1

• fix dfs-diffusers

  #71 opened a year ago by casperdcl
  0

• Github Actions to handle build and test of the containers

  #49 opened a year ago by filopedraz
  0

• Docker images are not caching in the correct manner

  #50 opened a year ago by filopedraz
  0