[HN Gopher] Mistral AI Valued at $2B
       ___________________________________________________________________
        
       Mistral AI Valued at $2B
        
       Author : marban
       Score  : 265 points
       Date   : 2023-12-10 18:25 UTC (4 hours ago)
        
 (HTM) web link (www.unite.ai)
 (TXT) w3m dump (www.unite.ai)
        
       | Racing0461 wrote:
       | With the new AI regulations the EU is going to adopt, how long
       | will mistral be paris based?
        
         | rolisz wrote:
         | Maybe the regulations will be Mistral shaped.
        
         | Barrin92 wrote:
         | there's nothing in the new AI regulations hindering Mistral's
         | work. Open Source foundation models are in no way impacted.
         | 
         | https://x.com/ylecun/status/1733481002234679685?s=20
        
           | Racing0461 wrote:
           | We both know that's not how regulations work. Mistral is
           | going to have to get a legal team to understand the
           | regulations, have a line item for each provision, verify each
           | one doesn't apply to them, get it signed off and continously
           | monitor for changes both to the laws and the code to make
           | sure it stays compliant. This will just be a mandate from
           | HR/Legal/Investors.
           | 
           | Alot of work for a company with no commercial offering off
           | the bat. And possibly an insurmountable amount of work for
           | new players trying to enter.
        
             | arlort wrote:
             | > Alot of work for a company with no commercial offering
             | off the bat
             | 
             | If you have no commercial offering it doesn't apply to you
             | at all in the first place
        
               | bsaul wrote:
               | If you never have any commercial offering, you have a 0
               | valuation.
        
         | andsoitis wrote:
         | Regardless of where a company is headquartered, it has to
         | comply with local regulations.
        
           | Racing0461 wrote:
           | Only if it wants to do business there. If a company is just
           | headquartered there, they have to comply with regulations no
           | matter what.
        
         | kozikow wrote:
         | Or another way to put it - if you are an enterprise based in
         | Europe that needs to stay compliant, future regulation will
         | make it very hard to not use Mistral :P.
        
       | matmulbro wrote:
       | LLM space is so cringe so much excitement from supply side and no
       | excitement/cringe from supposed demand side
        
         | huytersd wrote:
         | I don't know what you're talking about. I use chatGPT
         | extensively. Probably more than 50 times a day. I am extremely
         | excited for anything that can top the already amazing thing we
         | have now. They have a massive paying customer base.
        
           | 4death4 wrote:
           | What do you use it for?
        
             | dartos wrote:
             | I usually go to it before google now if I'm looking for an
             | answer to a specific question.
             | 
             | I know it can be wrong, but usually when it is, it's
             | obviously wrong
        
             | sjfjsjdjwvwvc wrote:
             | Not OP but I used it very successfully (not OpenAI but some
             | wrapper solution) for technical/developer support. Turns
             | out a lot of people prefer talking to a bot that gives a
             | direct answer than reading the docs.
             | 
             | Support workload on our Slack was reduced by 50-75% and the
             | output is steadily improving.
             | 
             | I wouldn't want to go back tbh.
        
             | kozikow wrote:
             | Not OP, but For me:
             | 
             | - Writing: emails, documentation, marketing - Write a bunch
             | of unstructured skeleton of information. Add a prompt about
             | the intended audience and a purpose. Possibly ask it to add
             | some detail.
             | 
             | - Coding: Especially things like "Is there a method for
             | this in this library" - a lot quicker than browsing through
             | documentation. Some errors - copy-paste the error from the
             | console, maybe a little bit for context, and quite often I
             | get the solution.
             | 
             | And API based:
             | 
             | - Support bot
             | 
             | - Prompt engineering of some text models that normally
             | would require labeling, training, and evaluation for weeks
             | or months. A couple of use cases - unstructured text as an
             | input + prompt, JSON as an output.
        
             | ipaddr wrote:
             | Bash scripts
        
             | s1artibartfast wrote:
             | I used it to write my wedding vows
        
               | huytersd wrote:
               | Based
        
             | huytersd wrote:
             | A lot of very varied things so it's hard to remember.
             | Yesterday I used it extensively to determine what I need to
             | buy for a chicken coop. Calculating the volume of concrete
             | and cinder blocks needed, the type and number of bags of
             | concrete I would need, calculating how many rolls of
             | chicken wire I would need, calculating the number of
             | shingles I would need, questions on techniques, and drying
             | times for using those things, calculating how much mortar I
             | would need for the cinderblocks (it took into account that
             | I would mortar only on the edges, the thickness of mortar
             | required for each joint, it accounted for the cores in the
             | cinderblocks, it correctly determined I wouldn't need
             | mortar on the horizontal axis on the bottom row) etc. All
             | of this, I could've done by hand, but I was able to sit and
             | literally use my voice to determine all of this in under
             | five minutes.
             | 
             | I use DALLE3 extensively for my woodworking hobby, where I
             | ask it to come up with ideas for different pieces of
             | furniture, and have constructed several based on those
             | suggestions.
             | 
             | For work I use it to write emails, to come up with
             | skeletons for performance reviews, look back look ahead
             | documents, ideas for what questions to bring up during
             | sprint reviews based on data points I provide it etc.
        
           | aantix wrote:
           | It's replaced Google for me, for most queries.
           | 
           | It's just so much more efficient in getting the answers I
           | need. And it makes a great pair programmer partner.
        
           | jay-barronville wrote:
           | 100%. ChatGPT is used heavily in my household (my wife and I
           | both have paid subscriptions) and it's absolutely worth it.
           | One of the most interesting things for me has actually been
           | watching my wife use it. She's an academic in the field of
           | education and I've seen her come up with so many creative
           | uses of the technology to help with her work. I'm a power
           | user too, but my usage, as a software engineer, is likely
           | more predictable and typical.
        
         | rogerkirkness wrote:
         | Microsoft Cloud AI revenue went $90M, $900M, $2.7B in three
         | quarters. How much more hard dollar demand growth could there
         | possibly be at this point?
        
           | matmulbro wrote:
           | it's shovels all the way down
        
             | sjfjsjdjwvwvc wrote:
             | shovelling what in your opinion? Or it's just a giant house
             | of cards?
        
               | cgearhart wrote:
               | Right now they're shoveling "potential". LLMs demonstrate
               | capabilities we haven't seen before, so there's high
               | uncertainty about the eventual impact. The pace of
               | progress makes it _seem_ like an LLM "killer app" could
               | appear any day and creating a sense of FOMO.
        
               | shrimpx wrote:
               | There's also the race to "AGI" -- companies spending tens
               | of billions on training, hoping they'll hit a major
               | intelligence breakthrough. If they don't hit anything
               | significant that would have been money (mostly) down the
               | drain, but Nvidia made out like a bandit.
        
             | quickthrower2 wrote:
             | I think there are enough genuine use cases. People are
             | saving time using AI tools. There are a lot of people in
             | office jobs. It is a huge market. Not to say it won't
             | overshoot. With high interest rates valuations should be
             | less frothy anyway.
        
           | echelon wrote:
           | They're selling to startups, not consumers.
           | 
           | The good startups are building, fine tuning, and running
           | models locally.
        
         | Xenoamorphous wrote:
         | I can't think of any software/service that's grown more in
         | terms of demand over a single year than ChatGPT (in all its
         | incarnations, like the MS Azure one).
        
         | itronitron wrote:
         | Yeah, the demand side consists solely of those that think they
         | will be supply side.
        
       | airspresso wrote:
       | Too many superlatives and groundbreaking miracles reported.
       | Probably written by AI.
        
         | jay-barronville wrote:
         | > In a significant development for the European artificial
         | intelligence sector, Paris-based startup Mistral AI has
         | achieved a noteworthy milestone. The company has successfully
         | secured a substantial investment of EUR450 million, propelling
         | its valuation to an impressive $2 billion.
         | 
         | I'm cracking up. I don't need to be a rocket scientist to read
         | this and immediately conclude it's AI-generated. I mean, they
         | didn't even try to hide that. Haha.
        
       | VirusNewbie wrote:
       | A competitor to OpenAI in like, benchmarks?
        
         | consumer451 wrote:
         | At least a competitor to Llama, for now.
         | 
         | https://medium.com/@datadrifters/mistral-7b-beats-llama-v2-1...
        
       | z7 wrote:
       | Mistral has a lot of potential, but there's the obvious risk that
       | without proper monetization strategies it might not achieve
       | sustainable profitability in the long term.
        
         | nothrowaways wrote:
         | Nothing stops them from launching a chat app.
        
           | quickthrower2 wrote:
           | The old open source, but we'll host it for you? I think Bezos
           | is going to be in fits of evil laughter about that model in 5
           | years, as all the open source compute moves to the clouds,
           | with dollars flowing his way.
           | 
           | But one thing Mistral could do is have a free foundational
           | model, and have non-free (as in beer, as in speech) "pro"
           | models. I think they will have to.
        
             | dartos wrote:
             | Release small, open, foundational models.
             | 
             | Deploy larger, fine tuned variants and charge for them.
             | 
             | There's a reason we don't have the data set or original
             | training scripts for mistral
        
               | behnamoh wrote:
               | it's a "mistry" ;)
        
             | teekert wrote:
             | Here's to hoping such models run on dedicated chips
             | locally, on Phones and PCs etc...
        
               | emadm wrote:
               | They already do, we just released a model equivalent to
               | most 40-60b base models that runs on a MacBook Air no
               | problem.
               | 
               | It's like 1.6gb, ones coming are better and smaller
               | https://x.com/EMostaque/status/1732912442282312099?s=20
               | 
               | I think the large language model paradigm is pretty much
               | done as we move to satisficing tbh
        
             | simonw wrote:
             | There are huge economy of scale benefits from providing
             | hosted models.
             | 
             | I've been trying out all sorts of open models, and some of
             | them are really impressive - but for my deployed web apps
             | I'm currently sticking with OpenAI, because the performance
             | and price I get from their API is generally much better
             | than I can get for open models.
             | 
             | If Mistral offered a hosted version which didn't have any
             | spin-up time and was price competitive with OpenAI I would
             | be much more likely to build against their models.
        
               | quickthrower2 wrote:
               | This only is defensible for closed models though.
        
           | echelon wrote:
           | Zero moat. Everybody's doing it.
           | 
           | I suppose they could be the Google to everyone else's Yahoo
           | and Dogpile, but I expect that to be a hard game to play
           | these days.
        
         | digitcatphd wrote:
         | I was wondering this. What is their business model exactly?
         | Almost seems like Europe's attempt to say "hey, look, we are
         | relevant too"
        
           | lolive wrote:
           | Being acquired.
        
         | skue wrote:
         | At this valuation and given the strength of the team, it's not
         | hard to imagine a future acquisition yielding a significant
         | ROI.
         | 
         | Besides, we don't know what future opportunities will unfold
         | for these technologies. Clearly there's no shortage of smart
         | investors happy to place bets on that uncertainty.
        
         | jsemrau wrote:
         | Model-as-a-service should work just fine.
        
         | stillwithit wrote:
         | Wait what? If company don't make $ it don't survive?
         | 
         | HN could really elevate the discourse if they flagged the
         | submarine ads of VCs
        
           | minimaxir wrote:
           | It is a relevant question in the AI industry specifically due
           | to new concerns about ROI given the intense compute costs.
        
             | lolive wrote:
             | Same concern I have regarding Spotify. [Which seems to have
             | insane recurring costs. Plus some risky expansive strategic
             | moves]
        
         | polygamous_bat wrote:
         | Coupled with the concern that once you're charging users money
         | for a product, you are also liable for sketchy things they do
         | with it. Not so much when you post a torrent link on twitter
         | that happens to have model weights.
        
         | niemandhier wrote:
         | The French have a urge to be independent, the French government
         | will hand them some juicy contract as soon as the can provide
         | any product that justifies that.
        
           | emadm wrote:
           | Yeah they shouldn't worry, they'll get a big French
           | government deal at worst
        
             | lolive wrote:
             | One of the French tycoons will eventually buy them.
        
           | yodsanklai wrote:
           | > The French have a urge to be independent
           | 
           | They lose that fight a long time ago though. It seems they
           | don't even try to pretend anymore.
        
         | dharma1 wrote:
         | On their pitch deck it said they will monetise serving of their
         | models.
         | 
         | While it may feel like a low moat if anyone can spin up a cloud
         | instance with the same model, it's still a reasonable starting
         | point. I think they will also be getting a lot of EU clients
         | who can't/don't want to use US providers.
        
       | foolfoolz wrote:
       | this is inevitable. at some point companies like this will be too
       | big to fail like air bus. maybe it's already there
        
       | jaspa99 wrote:
       | Curious to see how this will impact Aleph Alpha
        
         | emadm wrote:
         | Aleph Alpha raised even more ^_^
         | 
         | https://sifted.eu/articles/ai-startup-aleph-alpha-raises-500...
        
       | quickthrower2 wrote:
       | What is the business model?
        
         | hnarayanan wrote:
         | Sshh
        
           | quickthrower2 wrote:
           | Sorry I forgot, in AI $2Bn is preseed
        
         | malermeister wrote:
         | Get the French government to throw a ton of money at you for
         | sovereignty reasons
        
       | I_am_tiberius wrote:
       | I really hope that a European startup can successfully compete
       | with the major companies. I do not want to see privacy
       | violations, such as OpenAI's default use of user prompts for
       | training, become standard practice.
        
         | quickthrower2 wrote:
         | Does Anthropic count as European?
        
           | htrp wrote:
           | Dario is italian-american?
        
             | quickthrower2 wrote:
             | That doesn't matter too much, the corporate structure is
             | more interesting.
        
             | pb7 wrote:
             | Elon is South African but that doesn't make Tesla a South
             | African company.
        
           | uxp8u61q wrote:
           | How on Earth would it count as European? It's a completely
           | American company. Founded in the US, by Americans,
           | headquartered in the US, funded by American VCs... I
           | genuinely don't get how you arrived at the idea that it's
           | European.
        
             | quickthrower2 wrote:
             | Big office and lots of jobs in UK. And with complex tax
             | setups these days I wasn't sure.
        
               | uxp8u61q wrote:
               | By that measure I guess Apple is Irish...?!
        
               | totolouis wrote:
               | UK is not in the Europe anymore.
        
               | baal80spam wrote:
               | Interesting, TIL.
        
               | quickthrower2 wrote:
               | They cut through the continental shelf as part of Brexit.
        
               | denlekke wrote:
               | maybe not the distinction you meant but the UK is still
               | in Europe (the continent) and to me, European is a word
               | based on location not membership of the European Union
               | (which the UK left)
        
       | mark_l_watson wrote:
       | There is a lot of hype around LLMs, but (BUT!) Mistral well
       | deserves the hype. I use their original 7B model, as well as some
       | derived models, all the time. I can't wait to see what they
       | release next (which I expect to be a commercial product, although
       | the MoE model set they just released is free).
       | 
       | Another company worthy of some hype is 01.AI which released their
       | Yi-34B model. I have been running Yi locally on my Mac (use "
       | ollama run yi:34b") and it is amazing.
       | 
       | Hype away Mistral and 01.AI, hype away...
        
         | jay-barronville wrote:
         | You mind sharing what you find so amazing about Yi-34B? I
         | haven't had a chance to try it.
        
           | mark_l_watson wrote:
           | I just installed it on my 32B Mac yesterday, first
           | impressions: it does very well reasoning, it does very well
           | answering general common sense world knowledge questions, and
           | so far when it generates Python code, the code works and is
           | well documented. I know this is just subjective, but I have
           | been running a 30B model for a while in my Mac and Yi-34B
           | just feels much better. With 4bit quantization, I can still
           | run Emacs, terminal windows and a web browser with a few tabs
           | without seeing much page faulting. Anyway, please try it and
           | share a second opinion.
        
           | brucethemoose2 wrote:
           | The 200K finetunes are also quite good at understanding their
           | huge context.
        
         | dmos62 wrote:
         | How do you use these models? If you don't mind sharing. I use
         | GPT-4 as an alternative to googling, haven't yet found a reason
         | to switch to something else. I'll for example use it to learn
         | about the history, architecture, cultural context, etc of a
         | place when I'm visiting. I've found it very ergonomic for that.
        
           | teaearlgraycold wrote:
           | I've use lm studio. It's not reached peak user friendliness,
           | but it's a nice enough GUI. You'll need to fiddle with
           | resource allocation settings and select an optimally
           | quantized model for best performance. But you can do all that
           | in the UI.
        
           | risho wrote:
           | lm studio is an accessible simple way to use them. that said
           | expecting them to be anywhere near as good as gpt-4 is going
           | to lead to disappointment.
        
           | davidkunz wrote:
           | I use them in my editor with my plugin
           | https://github.com/David-Kunz/gen.nvim
        
             | 3abiton wrote:
             | Interesting use case, but the issue is wasting all this
             | compute energy for prediction?
        
               | HorizonXP wrote:
               | Can you explain what you mean by this question?
        
           | loufe wrote:
           | If you want to experiment Kobold.cpp is a great interface and
           | goes a long distance to guarantee backwards compatibility of
           | outdated model formats.
        
           | gdiamos wrote:
           | I host them here: https://app.lamini.ai/playground
           | 
           | You can play with them, tune them, and download the weights
           | 
           | It isn't exactly the same as open source because weights !=
           | source code, but it is close in the sense that it is editable
           | 
           | IMO we just don't have great tools for editing LLMs like we
           | do for code, but they are getting better
           | 
           | Prompt engineering, RAG, and finetuning/tuning are effective
           | for editing LLMs. They are getting easier and better tooling
           | is starting to emerge
        
         | p1esk wrote:
         | How do these small models compare to gpt4 for coding and
         | technical questions?
         | 
         | I noticed that gpt3.5 is practically useless to me (either
         | wrong or too generic), while gpt4 provides a decent answer 80%
         | of the time.
        
           | modeless wrote:
           | They are not close to GPT-4. Yet. But the rate of improvement
           | is higher than I expected. I think there will be open source
           | models at GPT-4 level that can run on consumer GPUs within a
           | year or two. Possibly requiring some new techniques that
           | haven't been invented yet. The rate of adoption of new
           | techniques that work is incredibly fast.
           | 
           | Of course, GPT-5 is expected soon, so there's a moving
           | target. And I can't see myself using GPT-4 much after GPT-5
           | is available, if it represents a significant improvement. We
           | are quite far from "good enough".
        
             | p1esk wrote:
             | I'm both excited and scared to think about this
             | "significant improvement" over GPT-4.
             | 
             | It can make our jobs a lot easier or it can take our jobs.
        
               | stavros wrote:
               | Isn't that the same? At some point, your job becomes so
               | easy that anyone can do it.
        
               | Spivak wrote:
               | It's weird for programmers to be worried about getting
               | automated out of a job when my job as a programmer is
               | basically to try as hard as I can to automate myself out
               | of a job.
        
               | rmbyrro wrote:
               | I expect the demand for SWE to grow faster than
               | productivity gains.
        
               | __loam wrote:
               | LLMs are going to spit out a lot of broken shit that
               | needs fixing. They're great at small context work but
               | full applications require more than they're capable of
               | imo.
        
             | OfSanguineFire wrote:
             | Curious thought: at some point a competitor's AI might
             | become so advanced, you can just ask it to tell you how to
             | create your own, analogous system. Easier than trying to
             | catch up on your own. Corporations will have to include
             | their own trade secrets among the things that AIs aren't
             | presently allowed to talk about like medical issues or sex.
        
               | p1esk wrote:
               | How to create my own LLM?
               | 
               | Step 1: get a billion dollars.
               | 
               | That's your main trade secret.
        
               | chongli wrote:
               | What is inherent about AIs that requires spending a
               | billion dollars?
               | 
               | Humans learn a lot of things from very little input.
               | Seems to me there's no reason, in principle, that AIs
               | could not do the same. We just haven't figured out how to
               | build them yet.
               | 
               | What we have right now, with LLMs, is a very crude brute-
               | force method. That suggests to me that we really don't
               | understand how cognition works, and much of this brute
               | computation is actually unnecessary.
        
               | nemothekid wrote:
               | If we knew how to build humans for cheap, then it
               | wouldn't require spending a billion dollars. Your
               | reasoning is circular.
               | 
               | It's precisely because we don't know how to build these
               | LLMs cheaply that one must so spend so much money to
               | build them.
        
               | chongli wrote:
               | The point is that it's not inherently necessary to spend
               | a billion dollars. We just haven't figured it out yet,
               | and it's not due to trade secrets.
               | 
               | Transistors used to cost a billion times more than they
               | do now [1]. Do you have any reason to suspect AIs to be
               | different?
               | 
               | [1] https://spectrum.ieee.org/how-much-did-early-
               | transistors-cos...
        
               | jryle70 wrote:
               | > Transistors used to cost a billion times more than they
               | do now
               | 
               | However you would still need billions of dollars if you
               | want state of the art chips today, say 3nm.
               | 
               | Similarly, LLM may at some point not require a billion
               | dollars, you may be able to get one, on par or surpass
               | GPT4, easily for cheap. The state of the art AI will
               | still require substantial investment.
        
               | pixl97 wrote:
               | >Humans learn a lot of things from very little input
               | 
               | And also takes 8 hours of sleep per day, and are mostly
               | worthless for the first 18 years. Oh, also they may tell
               | you to fuck off while they go on a 3000 mile nature walk
               | for 2 years because they like the idea of free love
               | better.
               | 
               | Knowing how birds fly ready doesn't make a useful
               | aircraft that can carry 50 tons of supplies, or one that
               | can go over the speed of sound.
               | 
               | This is the power of machines and bacteria. Throwing
               | massive numbers at the problem. Being able to solve
               | problems of cognition by throwing 1GW of power at it will
               | absolutely solve the problem of how our brain does it
               | with 20 watts in a faster period of time.
        
               | janalsncm wrote:
               | Because that billion dollars gets you the R&D to know how
               | to do it?
               | 
               | The original point was that an "AI" might become so
               | advanced that it would be able to describe how to create
               | a brain on a chip. This is flawed for two main reasons.
               | 
               | 1. The models we have today aren't able to do this. We
               | are able to model existing patterns fairly well but
               | making new discoveries is still out of reach.
               | 
               | 2. Any company capable of creating a model which had
               | singularity-like properties would discover them first,
               | simply by virtue of the fact that they have first access.
               | Then they would use their superior resources to write the
               | algorithm and train the next-gen model before you even
               | procured your first H100.
        
               | michaelt wrote:
               | Maybe not $1 billion, but you'd want quite a few million.
               | 
               | According to [1] a 70B model needs $1.7 million of GPU
               | time.
               | 
               | And when you spend that - you don't know if your model
               | will be a damp squib like Bard's original release. Or if
               | you've scraped the wrong stuff from the internet, and
               | you'll get shitty results because you didn't train on a
               | million pirated ebooks. Or if your competitors have a
               | multimodal model, and you really ought to be training on
               | images too.
               | 
               | So you'd want to be ready to spend $1.7 million more than
               | once.
               | 
               | You'll also probably want $$$$ to pay a bunch of humans
               | to choose between responses for human feedback to fine-
               | tune the results. And you can't use the cheapest workers
               | for that, if you need great english language skills and
               | want them to evaluate long responses.
               | 
               | And if you become successful, maybe you'll also want $$$$
               | for lawyers after you trained on all those pirated
               | ebooks.
               | 
               | And of course you'll need employees - the kind of
               | employees who are very much in demand right now.
               | 
               | You might not need _billions_ , but $10M would be a
               | shoestring budget.
               | 
               | [1]
               | https://twitter.com/moinnadeem/status/1681371166999707648
        
               | rmbyrro wrote:
               | It might work for fine-tuning an open model to a narrow
               | use case.
               | 
               | But creating a base model is out of reach. You need an
               | order of probably hundreds of millions of $$ (if not
               | billion) to get close to GPT 4.
        
               | Xenoamorphous wrote:
               | As someone who doesn't know much about how these models
               | work or are created I'd love to see some kind of
               | breakdown that shows what % of the power of GPT4 is due
               | to how it's modelled (layers or whatever) vs training
               | data and the computing resources associated with it.
        
               | janalsncm wrote:
               | The limiting factor isn't knowledge of how to do it, it
               | is GPU access and RLHF training data.
        
             | 0xDEF wrote:
             | >I think there will be open source models at GPT-4 level
             | that can run on consumer GPUs within a year or two.
             | 
             | There is indeed already open source models rivaling
             | ChatGPT-3.5 but GPT-4 is an order of magnitude better.
             | 
             | The sentiment that GPT-4 is going to be surpassed by open
             | source models soon is something I only notice on HN. Makes
             | me suspect people here haven't really tried the actual
             | GPT-4 but instead the various scammy services like Bing
             | that claim they are using GPT-4 under the hood when they
             | are clearly not.
        
               | rmbyrro wrote:
               | Makes me suspect you don't follow HN user base very
               | closely.
        
               | refulgentis wrote:
               | You're 100% right and I apologize that you're getting
               | downvoted, in solidarity I will eat downvotes with you.
               | 
               | HNs funny right now because LLMs are all over the front
               | page constantly, but there's a lot of HN "I am an expert
               | because I read comments sections" type behavior. So many
               | not even wrong comments that start from "I know LLaMa is
               | local and C++ is a programming language and I know
               | LLaMa.cpp is on GitHub and software improves and I've
               | heard of Mistral."
        
             | vitorgrs wrote:
             | I believe one of the problems that OSS models need to
             | solve, is... dataset. All of them lack a good and large
             | dataset.
             | 
             | And this is most noticiable if you ask anything that is not
             | in English-American-ish.
        
           | CSMastermind wrote:
           | Mistral's latest just released model is well below GPT-3 out
           | of the box. I've seen people speculate that with fine-tuning
           | and RLHF you could get GPT-3 like performance out of it but
           | it's still too early to tell.
           | 
           | I'm in agreement with you, I've been following this field for
           | a decade now and GPT-4 did seem to cross a magical threshold
           | for me where it was finally good enough to not just be a
           | curiosity but a real tool. I try to test every new model I
           | can get my hands on and it remains the only one to cross that
           | admittedly subjective threshold for me.
        
             | rmbyrro wrote:
             | Still, for a 7B model, this is quite impressive.
        
             | espadrine wrote:
             | > _Mistral 's latest just released model is well below
             | GPT-3 out of the box_
             | 
             | The early information I see implies it is above. Mind you,
             | that is mostly because GPT-3 was comparatively low: for
             | instance its 5-shot MMLU score was 43.9%, while Llama2 70B
             | 5-shot was 68.9%[0]. Early benchmarks[1] give Mixtral
             | scores above Llama2 70B on MMLU (and other benchmarks),
             | thus transitively, it seems likely to be above GPT-3.
             | 
             | Of course, GPT-3.5 has a 5-shot score of 70, and it is
             | unclear yet whether Mixtral is above or below, and clearly
             | it is below GPT-4's 86.5. The dust needs to settle, and the
             | official inference code needs to be released, before there
             | is certainty on its exact strength.
             | 
             | (It is also a base model, not a chat finetune; I see a lot
             | of people saying it is worse, simply because they interact
             | with it as if it was a chatbot.)
             | 
             | [0]: https://paperswithcode.com/sota/multi-task-language-
             | understa...
             | 
             | [1]: https://github.com/open-compass/MixtralKit#comparison-
             | with-o...
        
             | brucethemoose2 wrote:
             | Have you played with finetunes, like Cybertron? Augmented
             | in wrappers and retrievers like GPT is?
             | 
             | It's not there yet, but its waaaay closer than the plain
             | Mistral chat release.
        
           | idonotknowwhy wrote:
           | If you can run yi34b, you can run phind-codellama. It's much
           | better than yi and mistral for code questions. I use it
           | daily. More useful than gpt3 for coding, not as good as gpt4,
           | except that I can copy and paste secrets into it without
           | sending them to openai.
        
             | mark_l_watson wrote:
             | Thanks, I will give codellama a try.
        
           | sharemywin wrote:
           | what types of things do you ask ChatGPT to do for you
           | regarding coding?
        
           | valval wrote:
           | Open source models will probably catch up at the same rate as
           | open source search engines have caught up to Google search.
        
         | yodsanklai wrote:
         | > I use their original 7B model, as well as some derived
         | models, all the time.
         | 
         | How does it compare to other models? and with chatgpt in
         | particular?
        
           | valval wrote:
           | No comparison to be made.
        
         | brucethemoose2 wrote:
         | I concur, Yi 34B and Mistral 7B are fantastic.
         | 
         | But you need to run the top Yi finetunes instead of the vanilla
         | chat model. They are far better. I would recommend
         | Xaboros/Cybertron, or my own merge of several models on
         | huggingface if you want the long context Yi.
        
       | transformi wrote:
       | Evaluation based on what? what is the business model?
        
         | antirez wrote:
         | I believe that the rationale is that if you can do an
         | outstanding 7B model, it is likely that you are able to create,
         | in the near future, something that may compete with OpenAI, and
         | something that makes money, too.
        
       | minimaxir wrote:
       | Of course, the reason Mistral AI got a lot of press and publicity
       | in the first place was because they _open-sourced_ Mistral-7B
       | despite the not-making-money-in-the-short-term aspect of it.
       | 
       | It's better for the AI ecosystem as a whole to incentive AI
       | startups to make a business through good and open software
       | instead of building moats and lock-in ecosystems.
        
         | jeron wrote:
         | They ought to rename to "ReallyOpenAI"
        
         | sillysaurusx wrote:
         | I don't think that counts as open source. They didn't share any
         | details about their training, making it basically impossible to
         | replicate.
         | 
         | It's more akin to a SaaS company releasing a compiled binary
         | that usually runs on their server. Better than nothing, but not
         | exactly in the spirit of open source.
         | 
         | This doesn't seem like a pedantic distinction, but I suppose
         | it's up to the community to agree or disagree.
        
           | minimaxir wrote:
           | It's IMO a pedantic distinction.
           | 
           | A compiled binary is a bad metaphor because it gives the
           | implication that Mistral-7B is an as-is WYSIWIG project
           | that's not easily modifiable. In contrast, there have been a
           | bunch of new powerful new models created by modifying or
           | finetuning Mistral-7B such as Zephyr-7B:
           | https://huggingface.co/HuggingFaceH4/zephyr-7b-beta
           | 
           | The better analogy to Mistral-7B is something like modding
           | Minecraft or Skyrim: although those games are closed source
           | themselves, it has enabled innovations which helps the open-
           | source community directly.
           | 
           | It would be _nice_ to have fully open-source methodologies
           | but lacking them isn 't an inherent disqualifier.
        
             | hedgehog wrote:
             | It's a big distinction, if I want to tinker with the model
             | architecture I essentially can't because the training
             | pipeline is not public.
        
               | minimaxir wrote:
               | If you want to tinker with the architecture Hugging Face
               | has a FOSS implementation in transformers: https://github
               | .com/huggingface/transformers/blob/main/src/tr...
               | 
               | If you want to reproduce the _training pipeline_ , you
               | couldn't do that even if you wanted to because you don't
               | have access to thousands of A100s.
        
               | hedgehog wrote:
               | I'm well aware of the many open source architectures, and
               | the point stands. Models like GPT-J have open code and
               | data, and that allows using them as a baseline for
               | architecture experiments in a way that Mistral's models
               | can't be. Mistral publishes weights and code, but not the
               | training procedure or data. Not open.
        
               | sillysaurusx wrote:
               | We do, via TRC. Eleuther does too. I think it's a bad
               | idea to have a fatalistic attitude towards model
               | reproduction.
        
               | hedgehog wrote:
               | Exactly, nice work BTW. And no hate for Mistral, they're
               | doing great work, but let's not confuse weights-available
               | with fully open models.
        
               | emadm wrote:
               | With all the new national supercomputers scale isn't
               | really going to be an issue, they all want large language
               | models on 10k GH200s or whatever and the libraries are
               | getting easier to use
        
           | mrob wrote:
           | According to the Free Software Definition:
           | 
           | "Source code is defined as the preferred form of the program
           | for making changes in. Thus, whatever form a developer
           | changes to develop the program is the source code of that
           | developer's version."
           | 
           | According to the Open Source Definition:
           | 
           | "The source code must be the preferred form in which a
           | programmer would modify the program. Deliberately obfuscated
           | source code is not allowed. Intermediate forms such as the
           | output of a preprocessor or translator are not allowed."
           | 
           | LLM models are usually modified by changing the model weights
           | directly, instead of retraining the model from scratch. LLM
           | weights are poorly understood, but this is an unavoidable
           | side effect of the development methodology, not deliberate
           | obfuscation. "Intermediate" implies a form must undergo
           | further processing before it can be used, but LLM weights are
           | typically used directly. LLMs did not exist when these
           | definitions were written, so they aren't a perfect fit for
           | the terminology used, but there's a reasonable argument to be
           | made that LLM weights can qualify as "source code".
        
             | lmm wrote:
             | > LLM models are usually modified by changing the model
             | weights directly, instead of retraining the model from
             | scratch. LLM weights are poorly understood, but this is an
             | unavoidable side effect of the development methodology, not
             | deliberate obfuscation.
             | 
             | They're understood based on knowing the training process
             | though, and a developer working on them would want to have
             | the option of doing a partial or full retraining where
             | warranted.
        
         | seydor wrote:
         | also because their model is unconstrained/censored. and they
         | are commited to that according to what they say, they build it
         | so others can build on it. GPTs are not finished business and
         | hopefully the open source community with surpass the early
         | successes.
        
       | asim wrote:
       | I have realised just how meaningless valuations now are. As much
       | as we use them as a marker of success, you can find someone to
       | write the higher valuation ticket when it suits their agenda too
       | e.g the markup, the status signal, or just getting the deal done
       | ahead of your more rationale competitors in the investment
       | landscape. Now that's not to say Mistral isn't a valuable company
       | or that they aren't doing good work. It's just valuation markers
       | are meaningless and most of this capital raise in the AI space is
       | about offsetting the cloud/GPU spend. Might get downvoted to
       | death but watching valuation news feels like no news.
        
         | seydor wrote:
         | It's smoke. but where there is smoke, there is some level of
         | fire
        
           | jack_riminton wrote:
           | Not if it's a smoke machine
        
       | mytailorisrich wrote:
       | Perhaps someone can answer this: this is a one year old company.
       | Does this mean that barriers to entry are low and replication
       | relatively simple?
        
         | emadm wrote:
         | Main barrier right now is access to supercompute and how to run
         | it, everything is standardising quickly in the space
        
         | cavisne wrote:
         | The part of Meta research that worked on LLaMa happened to be
         | based in the Paris office. Then some of the leads left and
         | started Mistral.
         | 
         | Complex/simple is not really the right way to think about
         | training these models, I'd say its more arcane. Every mistake
         | is expensive because it takes a ton of GPU time and/or human
         | fine tuning time. Take a look at the logbooks of some of the
         | open source/research training runs.
         | 
         | So these engineers have some value as they've seen these
         | mistakes (paid for by Meta's budget).
        
       | JonChesterfield wrote:
       | Anyone else think Nvidia giving companies money to spend on
       | Nvidia hardware at very high profit margin is a dubious valuation
       | scheme?
        
         | raverbashing wrote:
         | You'd be surprised how this is much more common than people
         | realize
        
         | candiddevmike wrote:
         | It's the heads I win, tails you lose investment model
        
         | SeanAnderson wrote:
         | Why would it be a dubious valuation scheme? I guess if an
         | investor is looking at just revenue, or only looking at one
         | area of their business finances, maybe? Otherwise it seems like
         | the loss in funds would be weighed against the increase in
         | revenue and wouldn't distort earnings.
        
           | JonChesterfield wrote:
           | Say big green gives a company $100M with the rider that it
           | needs to spend all that on nvidia's hardware in exchange for
           | 10% of the company.
           | 
           | Has Nvidia valued the company at 1B? Say their margin is 80%
           | on the sales. So Nvidia has lost some cashflow and $20M for
           | that 10%. Has Nvidia valued the company at $200M?
        
             | SeanAnderson wrote:
             | I see :) Thanks for clarifying. I would say that I don't
             | have a strong enough grasp on biz finances to do more than
             | speculate here, but:
             | 
             | 1) Is all the money spent up front? Or does it trickle back
             | in over a few years? Cash flow might be impacted more than
             | implied, but I doubt this is much of an issue.
             | 
             | 2) I wonder how the 10% ownership at 2B valuation would be
             | interpreted by investors. If it's viewed as a fairly liquid
             | investment with low risk of depreciation then yeah, I could
             | see Nvidia's strategy being quite the way to pad numbers.
             | OTOH, the valuation could be seen as pure marketing fluff
             | and mostly written off by the markets until regulations and
             | profitability are firmly in place.
        
           | wongarsu wrote:
           | If it was a good valuation scheme, then Nvidia giving them
           | $100 million at a $2 billion valuation would mean that Nvidia
           | thinks the company is worth $2 billion. But if Mistral uses
           | that money to buy GPUs that Nvidia sells with 75% profit
           | margin, the deal is profitable for Nvidia even if they
           | believe the company is worth only $0.5 billion (since they
           | effectively get 75% of the investment back). And if this deal
           | fuels the wider LLM hype and leads other companies to spend
           | just $50 million more at Nvidia, this investment is
           | profitable for Nvidia even if Mistral had negative value.
        
             | emadm wrote:
             | With convertible debt and many of these rounds investors
             | get the first money out, so the first 450m would go to the
             | investors.
        
         | mcmcmc wrote:
         | Kinda like MS giving OpenAI all those Azure credits?
        
       | racoonista wrote:
       | Unfortunately, the EU also just passed some AI regulations. Not
       | sure how they impact Mistral's work, but just FWIW.
        
         | malermeister wrote:
         | Why is that an unfortunately? We need regulations to set the
         | rules of the game.
        
           | bsaul wrote:
           | we don't even know what AI is truely going to look like in 2
           | years, and 2 years ago nobody cared. Isn't it a bit too early
           | to regulate a field that's barely starting ?
        
       | b2bsaas00 wrote:
       | Anyone has example of products that made large use of LLM API
       | that could make economics sense to use self-hosted model
       | (Mistral, LLAMA)?
        
         | sroussey wrote:
         | Im working on embeddings database of my personal information,
         | and ability to query it. Just a privacy reason.
        
       | Frummy wrote:
       | That's fair given it's 50 times more difficult to use their model
        
       | fidotron wrote:
       | There is a lot of noise here suggesting it is too much, but
       | relative to the supposed SV unicorns of two years ago this looks
       | like an absolute steal.
        
         | yreg wrote:
         | The macroeconomic situation 2 years ago and now was wildly
         | different.
        
       | hn_throwaway_99 wrote:
       | Perhaps too much off-topic, but I hate how the press (and often
       | the startups themselves) focuses on the valuation number when a
       | company receives funding. As we've seen in very recent history,
       | those valuation numbers are at best a finger in the wind, and of
       | course a big capital intensive project like AI requires a
       | valuation that is at least a couple multiples of the investment,
       | even if it's all essentially based on hope.
       | 
       | I think it would make much more sense to focus on the "reality
       | side" of the transaction, e.g. "Mistral AI received a EUR450
       | million investment from top tech VC firms."
        
         | shrimpx wrote:
         | The valuation is meaningful in the sense of "Mistral sells
         | 22.5% of company to VC firms."
        
       | nojvek wrote:
       | Valuation means Jack shit for early stage startup. WeWork was
       | valued at $50B at its peak.
       | 
       | Until a company is consistently showing growth in revenue and a
       | path to sustainable profitability, valuation is essentially wild
       | speculation.
       | 
       | OpenAI is wildly unprofitable right now. The revenue they make is
       | through nice APIs.
       | 
       | What is Mistral's plan for profitability?
       | 
       | Right now stability AI is in dumps and looking for a buyer.
       | 
       | Only companies I see making money in AI are those who live like
       | cockroaches and very capital efficient. Midjourney and Comma.ai
       | come to mind.
       | 
       | Very much applaud them for open release of models and weights.
        
         | evantbyrne wrote:
         | Valuation matters quite a bit for continued funding.
        
           | hauget wrote:
           | His point is with regards to reaching & maintaining
           | profitability, not revenue spending.
        
             | evantbyrne wrote:
             | It's too early for Mistral to focus on revenue. These AI
             | companies are best thought of as moonshot projects.
        
           | toss1 wrote:
           | Yes, and it can matter in a very bad way if you need to
           | subsequently have a "down round" (more funding at a lower
           | valuation).
           | 
           | Initial high valuations mean the founders get a lot of
           | initial money giving up little stock. This can be awesome if
           | they become strongly cash-flow positive before they run out
           | of that much runway. But if not, they'll get crammed hard in
           | subsequent rounds.
           | 
           | The more key question is: how much funding did they raise at
           | that great valuation, and is it sufficient runway? Looks like
           | EUR450 million plus an additional EUR120 million in
           | convertible debt. Might be enough, depending on their
           | expenses...
        
             | evantbyrne wrote:
             | I'm not saying that either of your concerns are invalid.
             | The LLM space is just the wrong place to be for investors
             | who are worried about cash-flow positivity this early in
             | the game. These models are crazy expensive to develop
             | _currently_, but they is getting cheaper to train all the
             | time. Meaning Mistral spent a fraction of what OpenAI did
             | on GPT-3 to train their debut model, and that companies
             | started one year from now will be spending a fraction of
             | what both are spending presently to train their debut
             | models.
        
         | emadm wrote:
         | It's kinda weird thinking deep tech companies should be
         | profitable a year in.
         | 
         | Like it takes time to make lots of money and it's really hard
         | to build state of the art models.
         | 
         | Reality is this market is huge and growing massively as it is
         | so much more efficient to use these models than many (but not
         | all) tasks.
         | 
         | At stability I told team to focus on shipping models as next
         | year is the year for generative media where we are the leader
         | as language models go to the edge.
        
           | mpalmer wrote:
           | They didn't say that companies should be profitable at a year
           | in.
           | 
           | To my mind they just seemed to be responding to the slightly
           | clickbait-y title, which focuses on the valuation, which has
           | some significance but is still pretty abstract. Still,
           | headlines love the word "billion".
           | 
           | The straight-news version of the headline would probably
           | focus more on a16z's new round.
        
           | nojvek wrote:
           | I acknowledge it's easy to be an armchair critic. You are the
           | ones in battlefield doing real work and pushing the edge.
           | 
           | The thing is I don't want the pro-open-source players to
           | fizzle out and implode because funding dried up and they have
           | no path to self sustainability.
           | 
           | AGI could be 6 months away or 6 decades away.
           | 
           | E.g Cruise has a high probability of imploding. They raised
           | too much and didn't deliver. Now California has revoked their
           | license for driverless cars.
           | 
           | I'm 100% sure AGI, driverless cars and amazing robots will
           | come. Fairly convinced the ones who get us there will be the
           | cockroaches and not the dinosaurs.
        
             | emadm wrote:
             | I think its also tough at the early stage of the diffusion
             | (aha) of innovation curve, we are at the point of early
             | adopters and high churn before mass adoption of these
             | technologies over the coming years as they are good enough,
             | fast enough and cheap enough.
             | 
             | AGI is a bit of a canard imo, its not really actionable on
             | a business sense.
        
         | vagrantJin wrote:
         | comma.ai is a great example of a good business.
         | 
         | But I might have a bias because I was following along as the
         | company was built from whiteboard diagrams to what it became.
        
         | stavros wrote:
         | This is just tangential, but I wouldn't call their APIs "nice",
         | I'd be far less charitable. I spent a few hours (because that's
         | how long it took to figure out the API, due to almost zero
         | documentation) and wrote a nicer Python layer:
         | 
         | https://github.com/skorokithakis/ez-openai/
         | 
         | With all that money, I would have thought they'd be able to
         | design more user-friendly APIs. Maybe they could even ask an
         | LLM for help.
        
         | rmbyrro wrote:
         | Generally agree.
         | 
         | Instead of "path to profitability", I think path to ROI is more
         | appropriate, though.
         | 
         | WhatsApp never had a path to profitability, but it had a clear
         | path to ROI by building a unique and massive user base that
         | major social networks would fight for.
        
         | wslh wrote:
         | > OpenAI is wildly unprofitable right now.
         | 
         | Do we know some of its numbers? How many paid subscribers do
         | they have? I pay for two subscriptions.
        
         | segmondy wrote:
         | Profitability likewise means jack shit. You just need to be
         | have a successful acquisition by a lazy dinosaur or go make
         | enough income to go public. You can lose money for 10yrs
         | straight while transferring wealth from the public to the
         | investors/owners. With that said, I'm short Mistral for them
         | being French. I have absolute zero faith in EU based orgs.
         | 
         | On profitability, For all the new comers, I don't think anyone
         | can wager that any of them is going to make money. Capital
         | efficiency is overrated so long as they can survive for the
         | next year+, they are all trying to corner the market and OpenAI
         | is the one that seems to have found a way to milk the cow for
         | now. I truly believe that the true hitmakers are yet to enter
         | the scene.
        
       | wholien wrote:
       | how does Mistral monetize or plan to monetize? create a chat gpt-
       | like service and charge? license to other businesses?
        
       | nojvek wrote:
       | Gotta give it to Nvidia and TSMC. In the big AI race, they're the
       | ones with real moat and no serious competition.
       | 
       | No matter who wins, they'll need those sweet GPUs and fabs.
        
         | Yujf wrote:
         | Its the good old "in a gold rush, sell shovels"
        
       | ThalesX wrote:
       | My 1st thought as an European, "YAY! EU startup to the moon". My
       | 2nd thought was "n'aww, American VC". I guess that's the best we
       | can do around here.
        
         | paulddraper wrote:
         | It may feel that there are few EU startups and that's true.
         | 
         | But there are even fewer EU VCs.
        
           | ThalesX wrote:
           | Was CTO for some European startups. I'll always remember one
           | when by the time the EU VC was mid-way through its due
           | dilligence for 500k seed, we already had some millions lined
           | up from some US VCs no questions asked.
        
         | jamesblonde wrote:
         | The problem is that no European VC has that amount of capital.
         | European VCs typically have a couple of hundred million under
         | mgmt. SV VCs have a few billion under mgmt.
        
         | bsaul wrote:
         | There were european VCs investing in the very first round,
         | french one in particular. Founders are french. This qualifies
         | as european in my book (let's not get too demanding)
        
       | firebot wrote:
       | Who comes up with these valuations? The Donald?
        
       | eeasss wrote:
       | Some folks on this forum seem to get irritated by the prospect of
       | a successful AI company HQed in the EU. Why the hate?
        
       | yodsanklai wrote:
       | Noob questions (I don't know anything about LLM, I'm just a
       | casual user of ChatGPT)
       | 
       | - is what Mistral does better than Meta or OpenAI?
       | 
       | - will LLM become eventually open-source commodities with little
       | room for innovation or shall we expect to see a company with a
       | competitive advantage that will make it the new Google? in other
       | words, how much better can we expect these LLM to be in the
       | future? should we expect significant progress or have we reached
       | to diminished returns (after all, this is only statistical
       | prediction of next word, maybe there's an intrinsic limitation of
       | this method)
       | 
       | - are there some sorts of benchmarks to compare all these new
       | models?
        
       | nbzso wrote:
       | The old Masters have a saying: Never fall in love with your
       | creation. The AI industry is falling into the trap of their own
       | making (marketing). LLM's are nice toys, but implementation is
       | resource/energy expensive and murky at best. There are a lot of
       | real life problems that would be solved trough rational approach.
       | If someone is thirsty, the water is the most important part, not
       | the type of glass:)
        
         | TeMPOraL wrote:
         | If you compared the efficiency of steam engines during
         | industrial revolution with the ones used today, or power
         | generation from 100 years ago to that of now, or between just
         | about any chemical process, manufacturing method or
         | agricultural technique at its invention and now, you'd be
         | amazed by the difference. In some cases, the activity of today
         | was _several orders of magnitude more wasteful_ just 100 years
         | ago.
         | 
         | Or, I guess look at how size, energy use and speed of computer
         | hardware evolved over the past 70 years. Point is,
         | implementation being, right now, "resource/energy expensive and
         | murky at best" is how many very powerful inventions look at the
         | beginning.
         | 
         | > _If someone is thirsty, the water is the most important part,
         | not the type of glass:)_
         | 
         | Sure, except here, we're talking about one group selling a
         | glass imbued with breakthrough nanotech, allowing it to keep
         | the water at desired temperature indefinitely, and continuously
         | refill itself by sucking moisture out of the air. Sometimes,
         | the type glass may really matter, and then it's not surprising
         | many groups strive to be able to produce it.
        
           | nbzso wrote:
           | Don't fall in love with your creation, is not stop creating.
           | 
           | https://www.cell.com/joule/fulltext/S2542-4351(23)00365-3
        
       | qeternity wrote:
       | I see a lot of comments asking what or how people are using these
       | models for.
       | 
       | The promise of LLMs is _not_ in chatbots (imho). At scale, you
       | will not even realize you are interacting with a language model.
       | 
       | It just happens to be that the first, most boring, lowest hanging
       | fruit products that OAI, Anthropic, et al pump out are chatbots.
        
       ___________________________________________________________________
       (page generated 2023-12-10 23:01 UTC)