Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Leanstral 1.5

docs.mistral.ai

Post reply on HN
Print view View original

41–50 of 158 posts

Re: Leanstral 1.5

#41

Post by Mashimo »

Grimblewald wrote: post #33

Got curious, sign up, add money to account, try to use. Can't, it's a labs model. Fine, let's enable labs. Can't, unspecified error. Fine, lets contact customer support as instructed, can't no customer support, just a half-assed FAQ, that seems vibe-coded and searched poorly, totally irrelevant answers coming up for all queries tried. Then it hit me: If AI makes good customer support, then why does no AI company use…

Further down someone said the support is great and they respond within the day.

Re: Leanstral 1.5

#42

Post by c7b »

I'm not sure I understand the Weights policy. This site says the weights are Apache-licensed, suggesting it's open weights. But I can't find a download link. Their Huggingface profile seems to only provide an earlier snapshot [0]. Any pointers on whether/where we can or will be able to download the weights?

[0] https://huggingface.co/mistralai/Leanstral-2603

Re: Leanstral 1.5

#43

Post by barrenko »

psalaun wrote: post #38
doctorpangloss wrote: post #3

Real talk, does anyone use anything from Mistral because it performs the best, by whatever secular metric of your choosing? Or is it only used "because EU"? Just focus on answering the question. I wonder if anyone has observed it perform better on any objective metric in any rigorous setting.

I use it as my workhorse for coding and general chat questions, because it's good enough 80% of the time, and indeed it's french/european (with heavy US capital tho...). We complain too much about not having enough major competitors in the IT space, to not support a burgeoning one even if it's less powerful than SOTA labs

Well, if you're a taxpayer in EU you're already supporting it implicitly.

Re: Leanstral 1.5

#44

Post by saghm »

Grimblewald wrote: post #33

Got curious, sign up, add money to account, try to use. Can't, it's a labs model. Fine, let's enable labs. Can't, unspecified error. Fine, lets contact customer support as instructed, can't no customer support, just a half-assed FAQ, that seems vibe-coded and searched poorly, totally irrelevant answers coming up for all queries tried. Then it hit me: If AI makes good customer support, then why does no AI company use…

No one ever thought it made good customer support. It makes cheap customer support, and quite a lot of companies already have shitty customer support because they don't care about it being good, so they're thrilled to get to cut costs further.

It's "good" from the perspective of a company that's annoyed to have to spend money on actually fixing things.

Re: Leanstral 1.5

#47

Post by criticalfault »

Grimblewald wrote: post #33

Got curious, sign up, add money to account, try to use. Can't, it's a labs model. Fine, let's enable labs. Can't, unspecified error. Fine, lets contact customer support as instructed, can't no customer support, just a half-assed FAQ, that seems vibe-coded and searched poorly, totally irrelevant answers coming up for all queries tried. Then it hit me: If AI makes good customer support, then why does no AI company use…

These guys don't answer emails. Same for qwant.

Sample of two, but I'm assuming french companies don't like to being contacted n English.

Re: Leanstral 1.5

#48

Post by BartjeD »

Confiks wrote: post #17
doctorpangloss wrote: post #3

Real talk, does anyone use anything from Mistral because it performs the best, by whatever secular metric of your choosing? Or is it only used "because EU"? Just focus on answering the question. I wonder if anyone has observed it perform better on any objective metric in any rigorous setting.

For writing and languange learning it's very decent, especially Mistral Large. The pricing is very good too. I really like the consistently low time to first token and good token per second. Claude, especially in the past, would be very inconsistent, often with outages. Mistral mostly just always works and is very fast. Technical questions are unfortunately hit or miss. I'm lately pretty much always using a system pr…

Mistral doesn't have caching on batches. For me that meant they are 10x more expensive than Google.

I think its dumb.

Their support is hidden away in a chat bubble at the bottom. But they do respond promptly.

Its decent, but after switching to Google i wouldn't go back

Re: Leanstral 1.5

#49

Post by gregman1 »

Grimblewald wrote: post #33

Got curious, sign up, add money to account, try to use. Can't, it's a labs model. Fine, let's enable labs. Can't, unspecified error. Fine, lets contact customer support as instructed, can't no customer support, just a half-assed FAQ, that seems vibe-coded and searched poorly, totally irrelevant answers coming up for all queries tried. Then it hit me: If AI makes good customer support, then why does no AI company use…

“Don’t get high on your own supply” - I think it’s Microsoft’s motto.

Re: Leanstral 1.5

#50

Post by cavenditti »

refulgentis wrote: post #21
doctorpangloss wrote: post #3

Real talk, does anyone use anything from Mistral because it performs the best, by whatever secular metric of your choosing? Or is it only used "because EU"? Just focus on answering the question. I wonder if anyone has observed it perform better on any objective metric in any rigorous setting.

OCR is off the charts good on every metric you can think of. LLMs are a near-afterthought at this point if you don’t have data residency requirements. I love them and they’re slightly under rated, their models are consistently well-trained, open, but as you note, behind. There is no metric that will say they’re ahead in anything.

This. Best OCR provider by measure and it’s been for years
Post reply on HN