smallcloudai/refact

Issues

• Support Qwen2.5-Coder

  #457 opened 3 months ago by alexkramer98
  0

• EPIC: Add context-awareness on entire codebase

  #189 opened 3 months ago by klink
  0

• VSCode plugin broken by "Cannot reach the server:..."

  #379 opened 3 months ago by st01cs
  10

• Not working gpu filtering for Codellama/7b

  #423 opened 3 months ago by hazratisulton
  1

• how to fineture with codeLlama-7B

  #453 opened 3 months ago by deepforest7
  3

• ADD ability to use any model from any source

  #451 opened 5 months ago by allanlaal
  1

• ADD version number (and release date) to footer

  #450 opened 5 months ago by allanlaal
  1

• [VS Code] Multiline completion not working in some cases

  #315 opened 5 months ago by hazratisulton
  1

• Supporting the code model Codestral-22B-v0.1

  #435 opened 7 months ago by 596192804
  0

• VRAM memory leak for Refact.AI 1.6B

  #332 opened 8 months ago by tawek
  10

• [Feature Request] Add Slider to Change Context Size of Passthrough Models

  #426 opened 8 months ago by valaises
  0

• docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]].

  #410 opened 9 months ago by koenig-arthur
  1

• Add stablelm models

  #401 opened 9 months ago by valaises
  0

• Maybe hide popup when stats is empty?

  #389 opened 9 months ago by valaises
  2

• how add local model to mapping in docker-compse mount

  #397 opened 9 months ago by linpan
  0

• Missing link on page https://docs.refact.ai/faq/

  #383 opened 9 months ago by JohannesDienst-askui
  2

• Llama2 chat model times out

  #376 opened 10 months ago by jcntrl
  1

• stats problem

  #365 opened 10 months ago by valaises
  1

• run without database oss

  #366 opened 10 months ago by valaises
  1

• error running docker on wsl with cuda

  #368 opened 10 months ago by josersleal
  1

• LoRA metabug for v1.2

  #200 opened 10 months ago by olegklimov
  2

• Latest lora checkpoints for deepseek-coder/5.7b/mqa-base only generate 1 token to some requests

  #307 opened a year ago by hazratisulton
  1

• lora's "catastrophic forgetting" problem

  #311 opened a year ago by shatealaboxiaowang
  2

• Finetune Problem

  #235 opened a year ago by ChinnYu
  12

• Self host v1.4.0 MODEL always /infengine-v1/completions-wait-batch WAIT time out

  #300 opened a year ago by yourchanges
  10

• Support for HTTP proxies

  #207 opened a year ago by olegklimov
  0

• Simple API key for OSS version, so people can expose docker port via reverse proxy

  #232 opened a year ago by olegklimov
  1

• Finetune of deepseek-coder fails

  #262 opened a year ago by ryancu7
  7

• Finetune improvement for better performance

  #246 opened a year ago by hazratisulton
  2

• more files to process than processes

  #275 opened a year ago by assinchu
  4

• [ui] Finetune progress bar

  #202 opened a year ago by mitya52
  0

• Fix <th> number of columns in finetune

  #204 opened a year ago by mitya52
  0

• [CICL] cache flash_attn

  #223 opened a year ago by reymondzzzz
  0

• Host Model for Embeddings for RAG

  #239 opened a year ago by valaises
  0

• refact refuses to finetune if finds weird bytes in files

  #267 opened a year ago by worldemar
  2

• docker image fails to start on mac m3

  #257 opened a year ago by domdorn
  8

• Database not starting?

  #261 opened a year ago by m0ngr31
  2

• Finetune failed with "No train files provided"

  #242 opened a year ago by olegklimov
  0

• GPU Filtering improvement

  #244 opened a year ago by JegernOUTT
  0

• Self Hosted Chat Times Out VSCode

  #250 opened a year ago by stratus-ss
  0

• Add DeepSeek Coder models

  #216 opened a year ago by klink
  4

• Add Code LLaMA

  #217 opened a year ago by klink
  4

• add check for the minumum number of files for fine-tuning job

  #224 opened a year ago by klink
  0

• When Using Self-Hosted Server VS Code Plugin Does Not Send Requests

  #230 opened a year ago by artforlife
  2

• Refactoring of the finetuning script

  #219 opened a year ago by klink
  1

• Without DB, web UI should still open, tell what the problem is

  #227 opened a year ago by olegklimov
  0

• Could not scan repo with Refact/1.6B selected

  #218 opened a year ago by worldemar
  2

• EPIC: Run Self-hosted version on CPU

  #191 opened a year ago by klink
  6

• easy bug -- query_nvidia_smi doesn't have temp_celsius, ValueError: invalid literal for int() with base 10, "[N/A]"

  #203 opened a year ago by mitya52
  0

• Always "latest/best" now, instead of the clicked LoRA

  #186 opened a year ago by mitya52
  1