cfe32aa0a7
Central default for ChatModelId
2023-03-30 01:46:18 -07:00
3bcbf029b8
Cleanups
...
Thanks Harlan @harlanlewis
Input=password: inspired by https://github.com/enricoros/nextjs-chatgpt-app/commit/bc76a753ec4425986e4a1922ca0330ceaa668d22
Usage: https://github.com/enricoros/nextjs-chatgpt-app/commit/aefb2a4a850449f934176e170e7baa58915aa484
2023-03-30 01:39:48 -07:00
3532f4f9b6
Copy icons - closes #30
...
Icons show up when hovering. Tooltips galore.
2023-03-30 00:31:46 -07:00
baedf00cfd
Improve swapping purposes, and rendering
2023-03-29 22:49:08 -07:00
2eb4893727
Swappable Purpose, and cleanups
2023-03-29 18:07:31 -07:00
82da8e55e3
Double-click to Edit
2023-03-29 17:25:56 -07:00
2a11d1854f
Cleanup Chat, ChatMessage (vs Message), ChatMessageList (vs Conversation)
2023-03-29 15:26:37 -07:00
f16e744ef9
Smaller moves
2023-03-24 19:06:36 -07:00
2ea2de601a
Cleanups
2023-03-24 03:31:43 -07:00
ecdcdef749
Error message for Context size
2023-03-24 03:10:51 -07:00
65b8aab53e
Markdown: syntax highlight in code blocks
...
Implements part of #11 , where the markdown is in code blocks of this type:
```SomeName.md
# ...
```
2023-03-24 00:39:33 -07:00
f2bb334778
Markdown: syntax highlight in code blocks
...
Implements part of #11 , where the markdown is in code blocks of this type:
```markdown
# ...
```
2023-03-24 00:26:45 -07:00
2dafafdcf4
Improve running code. Can render basic HTML.
...
Fixes #19 . Now it's showing the HTML in the Sandpack box.
2023-03-23 23:33:52 -07:00
49bc3192ec
Explain model/system errors to the users
...
Closes #18 and closes #16 . Interesting find: the "code":"model_not_found" shows
that the model is just not authorized for this API key. If the model is really
non-existing and can't be found, the error code will be "null", and the message
will be "The model 'XYZ' does not exist".
2023-03-21 22:52:57 -07:00
ba8a281d80
Restart the conversation from any point
...
Also useful to try a few answers before moving forward. Good time saver.
2023-03-20 02:37:12 -07:00
2d0136c78f
Display the API model (as tooltip)
...
Behind the scenes, we detect the model name from the OpenAI
streaming API (it's in all the chunks), and stream it back
to the UI as a serialized JSON object. This is more of a
workaround, required by virtue of the streaming API, and can
be cleaned up later.
Model names available when hovering 'gpt-4' / '3.5-turbo'.
2023-03-19 23:08:38 -07:00
3d72246d72
Added support for GPT3.5 Turbo model
2023-03-19 14:18:53 -04:00
1677db1df8
Error Handling
...
Show the user if Errors are raised from the OpenAI API, or by
our backend. Good messaging, red coloring.
2023-03-19 05:11:17 -07:00
010b1b17cf
Run TS/JS/HTML blocks
...
Warning: this adds 203kb of JS to the app, more than 4x'ing
the size of the app. Will remove if not worth it. Uses Sandpack
from Codesandbox.io, which stands up full execution environments
for the code.
2023-03-19 00:59:30 -07:00
cd60c6cc27
Highlight Code blocks
...
Using prismjs, we add coloring of source code blocks, which
translates in 12k of additional code. But the eyes and mind
are so happy.
2023-03-19 00:53:52 -07:00
d2a7082564
Add Application
...
NextJS, React, JoyUI (like MUI), application for self-hosted
Chat-like GPT use cases. Includes streaming of responses from
the Vercel edge runtimes.
2023-03-19 00:42:23 -07:00
Enrico Ros