Skip to content
This repository was archived by the owner on Mar 9, 2026. It is now read-only.

Add Mistral provider to config template - #305

Merged
caarlos0 merged 1 commit into
charmbracelet:mainfrom
philippgille:add-mistral-config
Jul 29, 2024
Merged

caarlos0 merged 1 commit into
charmbracelet:mainfrom
philippgille:add-mistral-config

Conversation

@philippgille

@philippgille philippgille commented Jul 27, 2024 •

Copy link
Copy Markdown
Contributor

Mistral is known for their open models (originally Mistral 7b, then their Mixtral MoE model), but they also have proprietary models that they're offering through their inference API "La Plateforme": https://docs.mistral.ai/deployment/laplateforme/overview/

The original purely proprietary models (small, medium, large) are not SOTA anymore and might get deprecated at some point, but their latest model "Mistral Large 2" is supposedly close to Llama 3.1 405B (as per https://mistral.ai/news/mistral-large-2407/). (And while not entire proprietary, its weights can only be used under their non-commercial "Mistral Research License", while their platform allows commercial use)

The API is OpenAI-compatible, so mods works with it out-of-the-box.

@philippgille
philippgille requested a review from caarlos0 as a code owner July 27, 2024 13:50
@philippgille

Copy link
Copy Markdown
Contributor Author

Closes #174

@caarlos0
caarlos0 merged commit 0bf7d1c into charmbracelet:main Jul 29, 2024
@caarlos0

Copy link
Copy Markdown
Contributor

thanks!

@caarlos0 caarlos0 mentioned this pull request Jul 29, 2024
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants