Tom Dörr

Update from my side: The merge is more comprehensive than I thought, even after I solved all conflicts the tests don't pass. ``` ImportError: cannot import name 'OpenAIFunction' from partially...

``` $ nvidia-smi Fri Dec 8 17:41:43 2023 +---------------------------------------------------------------------------------------+ | NVIDIA-SMI 545.23.08 Driver Version: 545.23.08 CUDA Version: 12.3 | |-----------------------------------------+----------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr....

All the other libraries I tried didn't have batch processing (@mhillebrand). Using the hf pipeline seems pretty slow for some reason, time scales with batch size. ``` Pipeline - Batch...

Generation in the above case gets cut off after 32 tokens, so it should generate roughly the same number of tokens between the different generation modes. You are right that...

Using date-time might also help in time-performance analysis

Fixed it by doing ``` pip install -U aider-chat ```

Have the issue in kitty terminal as well as in the gnome terminal

Yes should also be supported by Kitty. Same issue with the `--no-pretty` flag:  Though it's not urgent, I'll just type out the filenames to avoid the issue

I'm starting it from Zsh which might be started from bash (Not quite sure how I configured it). Not using emacs, tmux or screen. 

Same result: