logoalt Hacker News

Microsoft AI: Seven new world-class models at scalable price-performance

173 points • by gmays • today at 4:08 PM • 114 comments • view on HN

Comments

johnshahbazian • today at 5:15 PM

I didn't watch the keynote, but I've been testing MAI Code 2 and its been quite good for my usual work. They put it on par with Opus or Sol, and I wouldn't argue too much with that. Kudos to them for starting from scratch and getting it right.

➕ show 1 reply
ahmedfromtunis • today at 4:38 PM

I think Microsoft should provide some of its models as open source/weight if they hope to secure any market share.

Otherwise, it's Windows Phone OS all over again.

➕ show 11 replies
M4R5H4LL • today at 4:53 PM

The Ignite keynote looked so shallow and opportunistic to leadership. You wonder if they have any clout anymore, and it feels like the engineers have abandoned the ship. I am working day to day with AI Engineering, and it feels I have zero reason to do business using any of their tech stack. This clearly started with the dev division re-org, they completely missed the memo. There is zero innovation, and it is more than time for Nadella to go.

➕ show 3 replies
theowaway213456 • today at 4:49 PM

Why are they not listed in the Artificial Analysis index?

reactordev • today at 6:06 PM

It’s about time Microsoft punched back. These are actually pretty good. Not just for a niche use case either (Phi).

jjyycc • today at 5:43 PM

I wonder if the cost is artificially lowered (using lower cost they set themselves for something) or if there are an objective calculation there.

➕ show 1 reply
tyingq • today at 5:29 PM

This seems a pretty bold statement, especially the Opus 5.5 comparison.

"MAI Code 2 reaches 98% of the coding quality of GPT-6.1 Sol at about 40% of the price. And against Opus 5.5, it reaches 96% performance for just 5% of the price."

➕ show 1 reply
cricketsandmops • today at 5:07 PM

I found their transcription model to be one of the best i have used

nickandbro • today at 4:34 PM

I think if they announced MAI Code 2 Flash to be not just for coding , but for other applications as well, it would do a lot better considering its 65% the cost of GPT-6 Luna.

➕ show 1 reply
caxco93 • today at 5:40 PM

so how do I use these exactly?

➕ show 2 replies
theanonymousone • today at 4:54 PM

Happy that this time they are releasing them beyond GHCP too. Looking forward to test the flash one.

gigel82 • today at 5:02 PM

MAI Image 2.6 Flash was about 2c / image, and it looks like 2.7 Flash is 1c / image. I used 2.6-Flash as a backup when my low cost Krea2 provider flaked out and the results were surprisingly ok (comparable with Krea2 Medium Turbo, cheapest provider cost is 0.67c / image).

At 1c it might even be a default choice, I'll definitely try it out.

nozzlegear • today at 4:51 PM

Release the weights you cowards!

ChrisArchitect • today at 5:17 PM

Related:

MAI Code 2 and Code 2 Flash

https://news.ycombinator.com/item?id=50535885

➕ show 1 reply
metalliqaz • today at 5:32 PM

is MAI a different team than the one that does Phi?

nhecker • today at 4:38 PM

[dead]

nik736 • today at 4:34 PM

"Decision 1.1 is the latest decision model from Microsoft, 50x faster and 79x cheaper than GPT-6.1 Sol"... uhm okay. What a useless comparison.

➕ show 7 replies
seizethecheese • today at 5:34 PM

> MAI Code 2 reaches 98% of the coding quality of GPT-6.1 Sol at about 40% of the price.

To me, this smells like the same trick I always see in claims about model routers. As in, running against a saturated benchmark then showing huge cost savings for a tiny performance drop (because almost any model will perform well).

Looking deeper, though, it looks like they are citing four benchmarks, only one of which is basically saturated (Terminal Bench 2.1).

On the other hand, on 2/4 of these benchmarks Sol 6.1 does better than Opus 5.5, which just doesn't pass the sniff test, so my conclusion is that these are cherrypicked benchmarks.

drewg123 • today at 6:11 PM

Amusing that this was announced 1 day before the Anthropic IPO

octoberfranklin • today at 6:12 PM

Closed weights, don't care. China's still beating you.

andai • today at 5:58 PM

>Security, alignment and containment are the most important priorities for agent development in 2027 and beyond.

Anyone else read John Lilly's stuff on SSI and laughed while reading that? Or at least, I imagine SSI is laughing ;)

999900000999 • today at 4:51 PM

> Today, we’re taking a big step forward with MAI Licensed Weights: models under your control.

>Take a frontier MAI model, customize it using your proprietary data if you want to. Then run it securely in your own Azure environment.

Ohh Microsoft. Is it really my environment if it’s Azure ?

Anyway I wonder if GitHub Co pilot will start auto routing to these new models. All those gamed efficiency metrics sure do look impressive.

This whole article is strange. It’s as if Microsoft is seeing the Anthropic IPO and saying “um we have models too”

➕ show 1 reply
behnamoh • today at 6:10 PM

> Decision 1.1 is the latest decision model from Microsoft, 50x faster and 79x cheaper than GPT-6.1 Sol.

No one compares decision models with Sol!

DeathArrow • today at 5:29 PM

>MAI Code 2 reaches 98% of the coding quality of GPT-6.1 Sol at about 40% of the price. And against Opus 5.5, it reaches 96% performance for just 5% of the price.

>MAI Code 2 Flash matches the performance of GPT-6 Luna at 65% of the price, and Haiku 5.5 at 20% of the price.

These are quiet some claims!

➕ show 1 reply
firemelt • today at 4:46 PM

and the crowd goes mild

➕ show 1 reply
j45 • today at 5:08 PM

Is it world class because someone says so?

➕ show 1 reply
neonihil • today at 5:47 PM

It will look like its useful. There will be business brochures about its being useful. Some sad engineers at sad megacorps will be forced to use it.

But at the end of the day: it will be what all other microsoft products turned up to be: an utter waste of time.

Spend your time on Claude, Gemini, Kimi, Trinity or even Deepseek. And thank me later.

ralph84 • today at 5:09 PM

"world-class" is one way to say "not-frontier"

➕ show 1 reply
mococa • today at 5:24 PM

No local models? Meh...

➕ show 1 reply
ahmedketata • today at 5:13 PM

[dead]

TutleCpt • today at 5:33 PM

Haven't used any Microsoft technologies for over 10 years. Why is this company still in business?

➕ show 3 replies
gimlichgu • today at 4:48 PM

Weak models, closed weights, no technical information

➕ show 1 reply