Mistral 7b instruct v0.3

[dapper solstice]

https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3

[loud elm]

https://api.together.ai/playground/chat/mistralai/Mistral-7B-Instruct-v0.3

[rough apex]

I hope it's the version Mistral had as "Mistral Tiny", it was quite a bit better than v0.2

[modern rock]

Should we route to this separately from the other versions?

I guess safest is to route separately for now then aliast later :-?...

[dapper solstice]

it has a totally different tokenizer it seems, sooo.... i guess?

[modern rock]

yeah

diff tokenizer

diff vocab

[buoyant needle]

I've been testing mistral q8_0 on my local machine and I really like it its really good for story writing/role playing

also for general q&a it seems really knowledgable and is uncensored

[buoyant needle]

https://www.youtube.com/watch?v=0T1444HJbt0

[gaunt palm]

According to https://docs.mistral.ai/getting-started/models/ the latest mistral-tiny is mistral-tiny-2312, but it doesn't say which version of Mistral-7B that refers to

[modern rock]

Almost there:

[rough apex]

btw there's apparently also an updated version of Mixtral-8x22B-Instruct out there
Idk if anyone hosted this yet
https://huggingface.co/mistral-community/mixtral-8x22B-Instruct-v0.3-original

[modern rock]

It's up: https://openrouter.ai/models/mistralai/mistral-7b-instruct-v0.3

Also https://openrouter.ai/models/mistralai/mistral-7b-instruct now points to 0.3

[loud elm]

https://deepinfra.com/mistralai/Mistral-7B-Instruct-v0.3