Move default model to gpt-4o-mini #91
Conversation
|
@PeterDaveHello Thank you 😄 |
| @@ -83,7 +83,7 @@ aicommit config set <key>=<value> | |||
| | `OPENAI_KEY` | N/A | The OpenAI API key. | | |||
| | `locale` | `en` | Locale for the generated commit messages. | | |||
| | `generate` | `1` | Number of commit messages to generate. | | |||
| | `model` | `gpt-3.5-turbo` | The Chat Completions model to use. | | |||
| | `model` | `gpt-4o-mini` | The Chat Completions model to use. | | |||
There was a problem hiding this comment.
@PeterDaveHello I think this model should be gpt-4o according to the current supported models
There was a problem hiding this comment.
I don't understand. It's quite clear:
As of July 2024, gpt-4o-mini should be used in place of gpt-3.5-turbo, as it is cheaper, more capable, multimodal, and just as fast. gpt-3.5-turbo is still available for use in the API.
There was a problem hiding this comment.
I completely agree with you - and i wish i could use the -mini , but as far as i can see it's only the -4o that is supported right now in their dictionary ?
Let me know if i'm misunderstanding something :)
There was a problem hiding this comment.
Is there any dependency on the repo you mentioned? I don't know why it's relevant.
There was a problem hiding this comment.
Yes - it's dependant on OpenAI's TikToken repo here: https://github.com/openai/tiktoken/blob/c0ba74c238d18b4824c25f3c27fc8698055b9a76/tiktoken/model.py#L22
There was a problem hiding this comment.
Maybe this is what you mean:
Line 32 in cad5e79
Then we should wait for dqbd/tiktoken#112. Otherwise, I don't understand why you asked me to open a PR here.
There was a problem hiding this comment.
Maybe using the official openai-node lib will prevent similar issues in the feature because they already supported gpt-4o-mini?
There was a problem hiding this comment.
@PeterDaveHello Good point. I will remove the dependency as it gives no value 😄
There was a problem hiding this comment.
If you update your branch now - i think it should work :)
@negoziator as you requested 😏