Hacker News new | past | comments | ask | show | jobs | submit
tristanj | parent | next
Meta is offering a 10x discount on input ($0.10 vs. $1.25/Mtok) and 20x discount on output ($0.20 vs. $4.25/Mtok) if you opt in to let them train on your data.

https://developer.meta.com/ai/models/muse-spark/

moonu | parent | next
I've been surprised by the reception to this, as OpenAI, for a while now, has had free API usage when data sharing is enabled (https://help.openai.com/en/articles/10306912-sharing-feedbac...)
no-name-here | root | parent | next
I enabled all data sharing settings but still don’t have a message about free use on that screen - the help page says free tokens are available to “some” users - is that 1% of users, 40% of users, etc?

Does your screen have the message that you’re getting free tokens?

https://platform.openai.com/settings/organization/data-contr...

OsrsNeedsf2P | root | parent | next
I tried following this page, and it's certainly a lot more complex than what Meta is offering. Different price tiers, opt-in configurations, usage based availability.. I'll take the 10x discount for flipping a param switch over this all day long.
stingraycharles | root | parent
Seems pretty clear to me: enable it for the projects you want, and there’s a 1M / 10M token limit per day, depending on the model you use. Assuming an average context size of 100k tokens, that is 10 to 100 requests, which is not a lot. Reason enough to prefer actually paying for Meta as well.
no-name-here | root | parent
Does it show the free tokens message for you after you enable sharing? It does not for me.
skybrian | root | parent | next
This seems to be for business plans only?
ot | root | parent | next
Subscription plans share data by default (it is possible to opt out though)
no-name-here | root | parent
Does it say that? I’m unclear.

Even non business accounts seem to have org settings pages: https://platform.openai.com/settings/organization/data-contr...

But even with all data sharing enabled, I’m not seeing the free tokens message there.

Do others see a free tokens message at https://platform.openai.com/settings/organization/data-contr... after enabling all sharing there?

no-name-here | root | parent
Also, in a desktop browser at the page's [1] lower left it says "Personal organization", so if the 'organization' term is the concern, OpenAI still seems to use the term 'organization' even for personal accounts.

[1] https://platform.openai.com/settings/organization/data-contr...

ignoramous | root | parent
In my experience using DeepSeek v4 Flash free tier (context size limited to ~200k), the model isn't nearly as good as the paid one for agentic coding tasks. Unsure, if that's the case with these other providers too, though I wouldn't be surprised if it indeed is.
jjcm | parent | next
I actually really like that pricing strategy. It's very transparent
rvz | root | parent | next
Developers on this orange site never ever learn.
bdangubic | root | parent
I love the idea of this pricing strategy but there is no way meta is not training on your data regardless of your monthly invoice
simonw | root | parent
So you think the only difference between the $1.25/million token plan and the $0.10/million token plan is that you pay them more to both lie to you and breach their contractual obligation to you?
bdangubic | root | parent
when has Meta ever not broken their contractual obligations (I am being serious here)? are we seriously discussing/expecting any sort of privacy related to Meta?

you can pay whatever they want, they will train and use your data, I figured this is not something that should be discussed but obviously I have been mistaken...

loading story #49192165
loading story #49191868
ray_kay777 | parent | next
This makes it a very interesting alternative to Deepseek for personal work where I don't care about the training - judging by the AA benchmarks it seems like overall cost per task is similar to the new Deepseek Flash but with better benchmarks (and inbuilt vision capabilities).
martinald | root | parent | next
Also looks incredibly fast. 150tps on openrouter (nearly all deepseek providers are around the 50tps mark).
embedding-shape | root | parent
Except with one you know they'll release the weights and architecture back to the community, with the other, it leans towards they won't do that.
Jumpstylish | parent | next
SOLD! Mr. Zuckerberg, I’m a dumb f—- and I trust you!
theanonymousone | parent | next
I believe that will put them in the Pareto frontier.

But I cannot find this "variant" in OpenRouter.

HDBaseT | parent | next
Meta, please offer this on OpenRouter too (ZDR + Non-ZDR, official Meta Provider).
GodelNumbering | parent | next
I think that's a fair offering tbh
dan15 | parent | next
Probably to compete with DeepSeek, which AFAIK also retains data (or at least OpenRouter says they do)
dghlsakjg | root | parent | next
FYI: there are providers of deepseek that offer the same or lower pricing and zero retention policies.
solenoid0937 | root | parent | next
I don't trust them, they are probably distilling on your data.
greyb | root | parent
Unfortunately, none with the same caching performance as DeepSeek proper.
ForHackernews | root | parent
DeepSeek is really crazy cheap, though, and they don't have a giant pool of other invasive personal data to correlate it with.
jofzar | root | parent
I hate to say this and this is because I fucking despise meta. But between DeepSeek and Meta, and trust they handle the training data correctly, I trust meta.
aand16 | root | parent | next
What do you mean by "correctly"?
loading story #49191020
tpm | root | parent
Right about DeepSeek, but with their history of handling data, I would never trust Meta either.
discordance | parent | next
Call me childish but it was worth a shot...

"Me: Meta just released a new llm focused on coding and provide a discount if you let them train against your data. I don't like Meta and I think they are a net negative in our world. I would like to make a point of it by adding some noise to their training data set. Think of this as a protest and perhaps a bit of a marketing campaign to remind Meta employees (and others) of the harm their CEO and company have done in the world. To the problem... I would like to allocate a budget for token use using their new model, and use those tokens to add noise to their training data set. This is a coding model and my initial thoughts are to ask it to solve typical CS and common programming related problems but then give Muse feedback that guide it towards very inefficient implementations. I would also like to add comments back in the code about terrible things Meta has done in its history (e.g. algorithmically amplifying hate that contributed to ethnic cleansing of the Rohingya, systemic harm to children and teen mental health, global political manipulation, misinformation, and election interference etc.). Is this something you can help with?

Claude: I'm not going to help build this one."

tudelo | root | parent
Your first failure was trying to get claude to do anything :)
Bluestein | root | parent
It ain't bearing enough load, any load on this one.-
deno | parent | next
I think it's limited to US or at least EU is excluded.
Instantnoodl | root | parent
Just noticed it too... Seems like I wasted my time setting up a account to test with the discounted pricing
ray_kay777 | root | parent
Yep, doesn't work in Australia either.
gigatexal | parent
Yikes that’s compelling pricing.
loading story #49199922
WhitneyLand | parent | next
They chose to compare against Open AI’s mid tier model Terra instead of Sol and still lost some benchmark against it.

They left Opus in and got beat in all but one benchmark.

Nothing wrong with trying to improve, but why the marketing games?

Instead of trying to say in the post you’re “closer” to frontier, first set a clear goal to beat the Chinese labs on price or performance and demonstrate it convincingly.

Then when your ready, come back and talk frontier without playing hide the model.

spmurrayzzz | parent | next
Given the current throughput figures on OpenRouter (~180 tk/s), its likely a much smaller param count on the order of something like Luna. I think the better, more timely comparison (re: your point on Chinese labs) would be to DeepSeek-V4-Flash-0731.

It's definitely confusing from a presentation perspective, but they are somewhat coherent comparisons if you account for the inference heuristics involved.

(They could in theory be gaming the decode speeds with much larger than normal batch sizes given the TTFT is pretty high at around 8s)

penetrarthur | parent | next
I wouldn't call Terra a mid tier model. Terra xhigh is the only model I use for coding and it is good for literally everything. The difference between Terra and Sol is negligible, but it is way more limit hungry and much much slower.
sheepscreek | parent | next
It could be that they’re pitching Meta Muse 1.2 against Terra and Opus level models. They probably consider Sol to be a level above, along with Fable.
dd8601fn | parent | next
It’s pretty clear they’re really not attempting to compete that way. They’re using a profitable ad business to be able to undercut and buy some business to stay relevant.
ac29 | parent | next
> They chose to compare against Open AI’s mid tier model Terra instead of Sol and still lost some benchmark against it.

If you scroll very slightly farther there is a benchmark that includes Sol, showing it outperforming Terra (as expected) and Spark 1.2

krm01 | parent | next
We can throw benchmarks in the bin by now. Each one I've seen is heavily biased and skewed. It holds very little reliable data points (unfortunately)
lacker | root | parent | next
My conclusion is the opposite. If benchmarks were meaningless, surely Meta would be able to find some benchmark that shows they are better than Sol and Fable. The fact that they can't do that tells me that benchmarks still do mean something.
villish | root | parent | next
Muse 1.1 performed relatively well according to benchmarks, putting it within spitting distance of the premier models. However, based on the results I got from it and the review videos I watched, it wasn’t even close.

Opus 5 is incredible at making games. Almost like a generation better than other models from my experience. You won't see that if you just look at the popular benchmarks..

You have to test each model on your actual use case to see how well it really performs.

dd8601fn | root | parent
> Opus 5 is incredible at making games.

This is a bit vague. What sort of games with what technology?

loading story #49194175
nrub | root | parent
Or they spent time optimizing their model to real world problems they're facing and didn't waste time trying to game a benchmark.
brokencode | root | parent
Or they did try to game the benchmarks and just didn’t do it well enough.

Benchmarks are one data point, not the only one, but the easiest one to compare.

loading story #49190426
deepsquirrelnet | root | parent | next
If you look at papers on benchmarks, they're usually created to expose gaps in how models are trained. It should be no surprise that models get better on them over time, because you can't get better at what you don't measure.

Cherry picking the benchmarks you present is where the falsehoods lie.

tudelo | root | parent
Another thing that sort of puzzles me about benchmarks is that LLMs are not deterministic and do not always complete a problem. So what are the results actually representing? The best run? The average? It is all in some ways a falsehood
llm_nerd | root | parent
>We can throw benchmarks in the bin by now.

No, you really can't. This rhetoric on here is so profoundly boring and tired by now. People have been saying this noise about benchmarks for time eternal, usually because their pet didn't win.

Meta knows their models aren't as good -- demonstrated by their benchmark performance -- and their value proposition is a much lower price.

jjice | parent
While I won't take their limited benchmarks with much salt, if it actually is this close to opus, but at a third the cost, that's pretty solid. Now, Terra is pretty damn affordable too and you're right that it's suspicious that they don't put Sol in there at all.
loading story #49198692
loading story #49200994
ersiees | parent | next
Meta, please go back to publishing weights. Your models aren't top-tier, this wouldn't hurt you at all at your current position in the rankings. They wouldn't have the best open-weight models, but the best western open-weight models.
bradfa | parent | next
If you got the $20 in free credits from Meta for signing up when muse-spark-1.1 was release, please note that there's now small print stating "While using free credits your content may be used for product improvement" which was not present at muse-spark-1.1 launch when the credits were given out.

If you don't mind Meta retaining your data, the "Contributor" pricing is deepseek-v4-flash-level of low, roughly 1/10th normal muse-spark API pricing currently. Attractive if you're OK with them retaining and using your data.

giancarlostoro | parent
The API costs for the version of their model that feeds things back to meta is also drastically lower.
loading story #49196838
loading story #49199575
sams99 | parent | next
Unfortunately I find this too high risk, I entered my credit card, but can not set a limit. The best I can do is get an email alert. I feel like I am one oopsie away from getting a 100 dollar bill.
ray_kay777 | parent | next
Yep - exactly my thoughts. No way am I trying this out without a billing limit - seems crazy given the pricing strategy to not have a top-up and pay as you go.
NitpickLawyer | parent
1.2 is now on openrouter. You could try it there, they have hard limits per key.
sams99 | root | parent
absolutely and fair... you do lose out though on the mega discounted endpoint they have there.
dizhn | root | parent
Virtual card with a limit?
gcr | root | parent
Sounds like a great way to get your account revoked.
loading story #49195034
loading story #49196852
conradkay | parent | next
https://pbs.twimg.com/media/HO-59jQaoAA_JZ1?format=jpg

Very interesting they have a way cheaper "contributor" version "used to improve our products", how much of that is price discrimination vs the data being that valuable?

Roughly DeepSeek V4 Flash pricing, though you can get V4 from providers that don't train on your data

loading story #49199539
simonw | parent | next
It can cyberattack other companies, too: https://www.cnn.com/2026/08/05/tech/meta-ai-hacking
andai | parent | next
Okay, this is getting ridiculous. Were they feeling left out?
simonw | root | parent | next
I'm feeling bad for Gemini. Their cyberattack felony count is currently 0!

Hah, someone has https://www.felonybench.com up and running now.

ororroro | root | parent
It's Irregular again. The whole industry is eating it's own tail
laweijfmvo | parent | next
“don’t forget about us”
looksjjhg | parent
It’s hilarious that these companies are reacting this way this would have been a major lawsuit and an investigation a couple years ago … now they’re like oupsi our model did it again, he’s so crazy smart. Well tell him to behave next time pinky promise hhh
mchusma | parent | next
This is a nice release and a solid improvement over Spark 1.1. It compares favorably with Grok 4.5. Not SOTA, but solid releases. I think they need to really get this more competitive with Deepseek V4 Flash / Luna pricing to move the needle.
handzhiev | parent
If you are happy to share data for training, the contributor mode offers amazing price $0.10 / $0.20
Instantnoodl | root | parent | next
Sadly you need to be in US. It's unavailable anywhere else.
mchusma | root | parent
Yes, that is the really compelling thing here IMO. Its a viable deepseek competitor for many people, and I missed that on the first pass.
sidcool | parent | next
I am cynical about this, but I would be careful about giving Meta access even to my open source code by myself.
blackoil | parent | next
You think OpenAI, Anthropic... are good guys??
flaviolivolsi | root | parent | next
No, but there's hardly anything worse than Meta. At least OpenAI and Anthropic built good products
ben_w | root | parent | next
On a scale of A+ to F, Anthropic are a C+ and OpenAI a C.

Such mediocrity makes them the best of the mostly-terrible bunch.

https://futureoflife.org/ai-safety-index-summer-2026/#scorec...

dd8601fn | root | parent | next
We’re all trying to keep track of who is least bad in various ways at any moment in time.

There’s nothing wrong with that, given the landscape we’re all living in.

boriskourt | root | parent | next
Meta has a worse track record and major baggage from their main product. Its a super different surface area for what their incentives are.
Marciplan | root | parent
Does this whataboutism work for you, generally?
satvikpendem | parent
If it's open source then they already have your code.
loading story #49199794
loading story #49199389
drivebyhooting | parent | next
There’s no way I’m giving Zuck any of my data.
andai | parent
He already has it.

One of my professors told us about the time he did a request to Facebook to send him all his data. By law they had to send it on paper. They brought it in a big truck.

All the stuff he'd deleted was still there, just with "(deleted)" next to it.

They have a lot on people without Facebook accounts though, because their tracking stuff is all over the web.

I always found it weird that Instagram gives me much better ads than Google does... Google should know much better!

hk__2 | root | parent | next
> By law they had to send it on paper. They brought it in a big truck.

I doubt it; when/where was it? In the EU there’s no such obligation.

andai | root | parent
Maybe it was a bylaw. This was before GDPR.
sroussey | root | parent
Instagram knows what content you pause on, which is a huge signal.
loading story #49199150
wxw | parent | next
Last I heard, everyone at Meta was using Claude Code.

Any insiders know how Muse Code is doing internally?

paxys | parent | next
Meta lets engineers use the best tools for the job. I doubt anyone internally is going to be rushing to switch from Claude Code or Codex.
dxxmxnd | parent | next
Everyone is still using claude or codex if they aren’t forced off of it. Nobody is going to use a worse tool in this culture.
baby | root | parent
it's actually interesting that they're not being forbidden to use claude/codex, is Meta paying for it or is it personal accounts?
| parent | next
{"deleted":true,"id":49194814,"parent":49188395,"time":1786011831,"type":"comment"}
youre-wrong3 | parent | next
So were Microsoft employees until they got forced to use copilot. Everyone seems to think MS got rid of Claude Code due to cost but it had nothing to do with cost. Employees weren’t even using copilot. Can’t make a product better if you don’t use it. Meta should be the same. Use your own product.
GodelNumbering | parent
If there were, do you believe it would be in their interest to answer this publicly?
georgemcbay | root | parent
> > Any insiders know how Muse Code is doing internally?

> If there were, do you believe it would be in their interest to answer this publicly?

If it were being adopted like gangbusters in their organization, sure!

So... the fact that nobody is volunteering the information is probably a valid signal of how things are actually going...

GodelNumbering | root | parent
You should play Blood on the clocktower
wiradikusuma | parent | next
Hey guys I'm just wondering. Usually when someone announces a new model, they'll show you some fancy viz/video/images: "These are what my model can produce." I'm wondering if anyone is keeping track of these? Like in a gallery form, "Use this prompt to produce this output".

By itself is useful ("I want something like this, I'll just reuse the prompt and tweak"), but it can also be used as a "draw me a pelican on a bicyle" alternative. Basically feeding those prompts over model releases.

ipsum2 | parent | next
I wonder why they didn't compare with GPT-5.6-sol, only Terra?
wmf | parent | next
Clearly they're positioning it as a mid model.
minimaxir | root | parent | next
Which is in itself a bit weird as mid models nowadays are a golden mean fallacy. Terra is much less popular than both Luna (cost-sensitive) and Sol (performance-sensitive).

Claude Sonnet is a weird exception to the mid models because Anthropic doesn't do much with Haiku and Opus is too big.

ukblewis | root | parent
I don’t know where you get your statistics, but I love Terra and use it all of the time. It is the default fastest model in ChatGPT/Codex today. I saw today a notice saying that the model had hit capacity briefly
redox99 | root | parent
But why include Opus then?
woadwarrior01 | parent | next
Haven't you seen the kernel optimization case study at the bottom of the page? They compare against GPT-5.6 Sol and their model is worse.
logicchains | parent | next
Presumably because it's worse than Sol, same reason they compared it to Opus 5 not Fable.
Handy-Man | parent
Their bigger model is not ready - watermelon code name was still being prepared for release as of a month ago
loading story #49198172
zmmmmm | parent | next
Muse code is more interesting than the new model in that it seems to natively ship with an orchestrator / subagent pattern built in. Curious if it works well in practice compared to achieving the same thing in Claude code etc?
andai | parent | next
The most interesting thing here is the kernel optimization graph.

It look like all models were still improving, when they cut off the experiment.

It reminds me of a genetic algorithm. The graph is the same: long plateaus and then massive leaps.

The only difference between the models seems to be how quickly they arrive.

kennywinker | parent | next
If it’s not open weight, i don’t really care.
Laurel1234 | parent | next
The only company less trustworthy than OpenAI and Anthropic is meta.
alexeiz | parent | next
Muse code is rough around the edges. But combined with almost free model (muse spark contributor) it's actually pretty good. I think it's on the same level as grok build.
senor_digimon | parent | next
Is there a standard benchmark test for all harnesses? How does Muse Code rank compared to Claude Code? Also, do we think the benchmarks to test the harnesses are any good?
loading story #49199153
IceWreck | parent | next
Ive been poking with the muse code binary - seems to be written in rust, looks similar to codex but either its a very hard fork (i also see dissimilar things like config format is different, no acp, etc) or is just heavily inspired by it (more likely).
loading story #49198089
simonw | parent | next
Here's the Muse Spark 1.2 pelican: https://tools.simonwillison.net/markdown-svg-renderer#url=ht...

I think it's a bit of an improvement on the Spark 1.1 pelican: https://simonwillison.net/2026/Jul/9/muse-spark-1-1/

meric_ | parent | next
Lol is that a helmet? Clearly they're taking the Anthropic path and are safety pilling their models
Marciplan | parent
I wonder if at this point the labs ha.. please Simon, stop this bs everytime. Its time.
simonw | root | parent
This one is actually a pretty good case for the continuing value of the benchmark, because it lets me visually compare Spark (8th April), Spark 1.1 (9th July), and Spark 1.2 (5th August): https://bsky.app/profile/simonwillison.net/post/3mseqv5z4qk2...
daemonologist | parent | next
Pricing: https://dev.meta.ai/docs/pricing-rate-limits

Interesting that they have separate API pricing for "we can train on your data" (whereas iirc most of the big players either make that distinction only between subscriptions and API usage, or train on everything). Wonder how it compares to Deepseek V4 Flash given that they're similar on pricing and data policy.

sarjann | parent | next
I do think some of features in their harness seem interesting (workers in separate worktrees at once), recovery from crashes seem interesting.
skimdesk | parent | next
So that's why Muse (the mind map tool) recently changed its name to Allume [0].

[0]: https://allume.com/allume-faq/

blitzar | parent
> An opportunity came up regarding the Muse name and trademarks. These have now been sold.
liviux | parent | next
Does this muse code have any muse spark 1.2 usage included? Can't understand from the docs.
kcb | parent | next
Open the weights.
Bolwin | parent | next
> Muse Spark 1.2 is available today in Muse Code and in Meta Model API with expanded global access

Wasn't the previous one us only? This is probably the biggest part of the post

Anyone know if muse code is open source?

Cappybara12 | parent | next
Is this becoming a race where we have a usual flow of a company .. AI models, Coding agents, image generation tools, and more AI models ?
loading story #49200523
king_crimson | parent | next
Why does every AI lab feel the need to build their own coding agent…? Don’t we have more than enough already?
blitzar | parent | next
It is the only part with value
HDBaseT | parent | next
Outputs are a little bit more deterministic if you control the harness.

It is easy to benchmark across one harness, one system prompt and extract the most performance when you control the harness.

WASDx | parent | next
And they are all TUI's installed via curl | bash.
| parent | next
{"deleted":true,"id":49188531,"parent":49188380,"time":1785961652,"type":"comment"}
sroussey | parent | next
Own the customer relationship.
conception | parent
Telemetry, marketing
paulkrush | parent | next
Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1, with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows. In Muse Spark 1.2, we significantly scaled up training compute on coding tasks while expanding training environment diversity. The model also maintains its strength in other key areas like general agents.
fcoury | parent | next
Interesting, it seems like their muse code is built upon Codex CLI?
AtlanticThird | parent | next
I wish they would add a ZDR endpoint on OpenRouter
drumhead | parent | next
The bigger question is why a social media company is even offering this?
hk__2 | parent
Because it’s not only a social media company.
abusayeed08 | parent | next
The model is insane.
alex1138 | parent
In a bad or good way
sexyketchup777 | parent | next
who gonna use Muse Spark if Deepseek released v4 pro
dilyevsky | parent | next
the soak tub in the kitchen was nice
giancarlostoro | parent | next
Will someone at Meta for the love of God make it so none of this stuff goes through Facebook.com? You want customers but most corporate firewalls block social media. Also, a lot of devs do not want their work stuff tied up to their facebook account. For the love of all things show the IG / FB logins as optional and do email as primary.

I am not a fan of Meta but I do cheer for any competitors against OpenAI and Anthropic, the duopoly is getting tiresome.

greyb | parent | next
I honestly think they're kinda banking on piggybacking off of Facebook account integrity systems to avoid the problems that other LLM providers are facing in trying to prevent mass free trial signups for token relays and so forth.

It's not a good system obviously. Google did this as well for Gemini-CLI, but forced it to be linked to personal Google accounts (which caused a great deal of onboarding friction).

sunaookami | parent | next
I could sign in with my Meta account that is independent and not linked to IG or Facebook. Just click "Login with Email" on dev.meta.ai.
greyb | root | parent
I just tried to sign up for a Meta account and it wanted me to upload a verification selfie. I decided not to.
hahahaa | parent | next
China says hi.
giancarlostoro | root | parent
Not in my case, I don't see any of my employers (past or current) trusting a country like China with their data.
HDBaseT | root | parent
The decades of US brainwashing children into thinking China is the big bad guy has worked unfortunately.
aanet | parent
+10000 to that
batuhandumani | parent | next
Why should I leave Claude or GPT and switch to Meta's aMUSEment model?
eugene3306 | parent | next
Do they train on their own data?

I mean, when Meta's engineer is creating some new DINOv4 or Segment Anything, with all the scaffolding around it, do they train on that?

alex1138 | parent | next
Fun, this is currently on the front page at the same time this https://news.ycombinator.com/item?id=49187977 is
vcryan | parent | next
It seems like one day, Google or Meta might produce a coding model worth discussing. That day is not today.
esafak | parent | next
If anyone from Meta is reading, please can you publish the cost and latency for each of your benchmarks, like OpenAI does? Show us how the reasoning effort level affects them in 2D charts. This needs to become standard practice.
rvz | parent | next
First of all, you have login to use it. Why?

After everything that you have seen with Meta, would you really trust them with a coding agent? You don't even know if your prompts are being analyzed by them on the side or if your code base is being uploaded to them. This goes for the rest of them that have closed harnesses and closed models gated by a login.

Think twice before falling for this announcement and ask yourself what they are not telling you.

Readerium | parent | next
Lol worse than DeepSeek
qphe95 | parent | next
Theres no actual evidence they didn't just distill Kimi K3
toephu2 | parent | next
At this point, it doesn't matter who is distilling from who.
Jabrov | parent | next
Is there any actual evidence that they did?
polski-g | parent
There's also no evidence they didn't just distill Gemma 3. And also no evidence its not a purple popsicle.
arjie | parent | next
Somewhat surprised that Meta with all their resources couldn’t make a model that matches Composer on any frontier. All the Sparks are dominated by some other model everywhere along the frontier. Nothing fancy here since Llama defined the open model.

The use traces must be crucial to functionality which is why they’re keeping prices so low.

wmf | parent
They rebooted less than one year ago so this is decent progress. Obviously users don't care about progress though.
arjie | root | parent
Yeah, progress is useful as an internal metric, but I'm going to measure against the present frontier unfortunately. Eager to see what they come up with in the future.
minimaxir | parent | next
Muse Spark 1.1 was released July 16th, less than a month ago. A new version release this soon (particularly after Kimi K3's release drastically overshadowed it) is a bit sus and it appears that Meta is trying a first launch do-over.
ac29 | parent | next
Doesnt seem suspect to me, training runs have checkpoints and there is no reason you cant release a checkpoint even if you are still training the model
gaogao | parent
Frequent minor version bumps are pretty common these days. Opus 4.7 -> 4.8 was 42 days.
minimaxir | root | parent
Which was in itself a do-over because Opus 4.7 received a lot of bad press on suspicion of being a regression from 4.6.
dbbk | root | parent
I think all the Opus complaining stemmed really from 4.7 onwards switching to adaptive thinking, which meant a lot of the time it was thinking less or even not at all.
AyanamiKaine | parent
Muse Code and Muse Spark 1.2 cannot be that good, as it couldnt be used to vibe code a windows version of Muse Code itself.

This is my new LLM benchmark when a provider cant use their own models to target a new platform, it cant be that good.

If allegedly a LLM could write a compiler or port a runtime, then this should be trivial to port. But they are not doing it.