Top
Best
New

Posted by kuberwastaken 5 hours ago

Mistral raises €3B(mistral.ai)
454 points | 313 commentspage 2
whazor 2 hours ago|
It makes no sense to train frontier models from scratch anymore. The best frontier models are only a half year ahead of Chinese open models. In this regard Anthropic and OpenAI are also in a bad spot when they waste so much compute on training models.

An important factor is that fine tuning existing open models is incredible cheap. You can easily change any cultural biases if you want a model to be 'sovereign'. And Mistral could combine that with their custom data sets for their enterprise customer needs. Mistral still trains their own models, but they also seem to offer fine tuning existing models.

With model weights being commoditized, another differentiator could be deploying efficient inference chips, especially if you combine it with a developer ecosystem for vendor lock-in. That is why it is interesting that both Samsung and ASML are investors, since they are companies that could make a difference in this area.

rhdunn 2 hours ago|
It only makes sense to train a frontier model if you are trying a different architecture to one that is available from an existing frontier model. This is because the different model architecture will learn the weights differently.

It may make sense to train a frontier model on an existing architecture if the base model is not available and the instruction trained version doesn't fit with what you want. There are techniques like ablation, but those could have other effects on the model, and there can still be lingering effects of the instruction training in the model that surface less frequently (e.g. on an input not covered by the ablation training).

Otherwise, fine tuning is definitely the way to go. However, you need to be careful not to over-tune the model such that it is only tuned to the data you are training it on.

steinvakt2 4 hours ago||
That’s great, but meanwhile American competitors are worth… trillions?
Daunk 4 hours ago||
*losing trillions
bpodgursky 4 hours ago||
Anthropic is thought to be cash flow positive now.
Systemerror7A69 4 hours ago|||
And my wallet is thought to contain a billion dollars, so long as I don't open it.

Seriously "thought to be" is such a baseless statement. Thought to be by whom? And on what basis?

benjaminwootton 3 hours ago||
It’s thought to be because it’s a private company but it’s very well cited.
nesk_ 3 hours ago||||
If they were cash flow positive they would tell it to everyone, be sure of that.
tristanj 2 hours ago||
Anthropic is currently in an IPO quiet period and cannot publish such information.
eloisant 3 hours ago||||
it's fake cash flow based on circular economy
adventured 35 minutes ago||
Billions of long-term users for the most important technology in human history isn't a premise of fake cash in a circular economy.

You might as well pretend there's no money in search or social media.

1999. That Google, it has no serious business model, it'll never make real money. Hello GoTo.

2006. That Facebook, it has no serious business model, it'll never make real money.

One more: 2020. That Uber, I could clone an MVP and kill them over the weekend if I really wanted to. That's a side project at best. They're bleeding to death anyway, losing money hand over fist, they'll never make it. They have no moat at all.

mirekrusin 4 hours ago|||
[dead]
hobofan 4 hours ago|||
Are they, though? I would say, let's find out when the IPO settles, but then again the market also sustains a incredibly crazy valuation for Tesla.
steinvakt2 4 hours ago||
If not, I’m buying stocks
nicce 4 hours ago||
Better to buy GPUs and RAM with current rate
danmaz74 4 hours ago|||
At this stage it's not about how much it's worth, but how much investment is required to reach their goals.
shikck200 4 hours ago||
Nah. They are not, everyone is losing money, and just staying alive from gov subsidies. A claude PRO license is 20 usd/month, and for it to be profitable it should be somewhere around 200-300/month.

The bubble is going to burst soon.

skeledrew 2 hours ago||
I think the Pro and other subs allow them to save big on ads (Claude Code use is required and it pushes their ads) and allow easy access to free data (CC occasionally solicits feedback and session sharing, which I'm pretty sure some users oblige). Also CC does massive prompt caching tailored to work in lockstep with their platform. With all that and who knows what else, I'd say it balances over time.
JanTurnherr 2 hours ago||
I'm using Mistral for a local VN Proofreader and its by far the best Local 14B Model for this specific use case.
alex_duf 1 hour ago|
What's VN in this context?
JanTurnherr 1 hour ago||
A Visual Novel. Its from a Software-Engineering standpoint interesting, since the requirements for this specific Use-Case are immense.
Tepix 2 hours ago||
I really hope that Mistral will catch up. I'm left wondering why so many Chinese labs manage to be competitive.

Perhaps Mistral has too high moral standards to keep up?

baq 3 hours ago||
10-20x more still needed, and then somewhere to build a couple DCs. Fingers crossed they make it, neither US nor Chinese labs can be trusted, even with open weights.
piyush_patel_ha 2 hours ago||
I have used some of models of mistral, I'm liking where they're heading.
AidenVennis 4 hours ago||
I really do hope they can catch up with the American models, but I think setting the Chinese models as the goal would be best. They seems to be able to create great models with low cost that are probably useful for 90% of the day to day tasks. Mistral should not focus on competing with Claude Fable or OpenAI's Astra at first but have a good EU alternative to Opus or even Sonnet. The fact that it's European will be enough to be used by a lot of companies and governments in the EU that are (trying to) move away from US tech.
Traubenfuchs 3 hours ago|
> but have a good EU alternative to Opus or even Sonnet

They would if they could, which means they can‘t, even though they want to.

glimshe 2 hours ago||
I know a lot of people who use AI, but literally nobody who uses Mistral.
0x20cowboy 4 hours ago||
Très bien! C’est parti!
darkteflon 3 hours ago|
Surely outfits like Sakana will be looking at a similar strategy. The White House / Anthropic spat supercharged all this.
More comments...