Commit History

Author SHA1 Message Date
  Alex Cheema 98079ab382 bt bench results 1 year ago
  Alex Cheema cedc52484f optimise bluetooth 1 year ago
  Alex Cheema e57e8caded bluetooth bench server 1 year ago
  Alex Cheema 50edb72079 bluetooth bench optimised characteristics 1 year ago
  Alex Cheema fd800f479b bluetooth bench list all services and their characteristics 1 year ago
  Alex Cheema aaf78f16cb optimised bluetooth bench 1 year ago
  Alex Cheema 0ffb1827b8 bluetooth bench use uuids 1 year ago
  Alex Cheema c931a0dfee bluetooth bench 1 year ago
  Alex Cheema 57215041a0 todo for speculative model 1 year ago
  Alex Cheema dc3b2bde39 use NousResearch/Meta-Llama-3.1-70B-Instruct as tinygrad llama-3.1-70b model, previously using non-instruct model 1 year ago
  Alex Cheema 5a9f4ba5c1 update examples: remove old llama3_distributed, add chatgpt_api 1 year ago
  Alex Cheema 581856897a clean up unused, formatting 1 year ago
  Alex Cheema 62e3726263 add RTX 20 series to device capabilities 1 year ago
  Alex Cheema ebff636a25 script ot start openwebui 1 year ago
  Alex Cheema 394935711b add all chat endpoints without v1 prefix to support ollama / openwebui. related: #175 1 year ago
  Alex Cheema 70172d7cb9 add /v1/models endpoint and change Content-Type of stremed response to text/event-stream. fixes #175 1 year ago
  Alex Cheema d917778e2b update mlx to 0.17.1 (not sure where 0.17.0 went on PyPi disappeared)g 1 year ago
  Alex Cheema 2667c8af44 cleaner download_progress 1 year ago
  Alex Cheema f46d077beb fix font dependencies for tinychat. related: #172 1 year ago
  Alex Cheema 8a4928f80c fix gitignore to not ignore tinychat static files 1 year ago
  Alex Cheema d515d9efa3 explicitly use absolute paths for tinychat deps 1 year ago
  Alex Cheema a386c35fde script to update tinychat deps 1 year ago
  Alex Cheema 3791e669a4 download tinychat dependencies all to local dir so we dont need internet 1 year ago
  Alex Cheema 85bab25ac0 fix local check if dir does not exist 1 year ago
  Alex Cheema 59c4393d95 first try loading tokenizer from local path instead of always going to the internet first. significant speed ups 1 year ago
  Alex Cheema 784e6bae21 print traceback on topology collection error 1 year ago
  Alex Cheema 8cad0e1849 only use_fast tokenizer for Mistral Large until this inconsistency bug is fixed #171 1 year ago
  Alex Cheema 85279007b3 hotfix edge case where we try to render before tokenizer is set 1 year ago
  Alex Cheema 09a8468395 upgrade mlx to 0.17.0 1 year ago
  Alex Cheema 1f9d16ec78 run tokenizers test in ci, run all models available 1 year ago