MirAI News

AI news for Power-users! Fastest AI news since owner doesn't sleep much | Owner: @oxfrancesco_

[MirAI News @MirAI_Newz]

Those are the best CURRENT Al models, let's see how much time it takes before one of those gets dethroned.

Follow to always be Updated!

[MirAI News @MirAI_Newz]

Those are the best CURRENT Al models, let's see how much time it takes before one of those gets dethroned.

Follow to always be Updated!

[MirAI News @MirAI_Newz]

This is Google's Decision API! Their direct answer to Jev's kind of model!

[Logan Kilpatrick @OfficialLoganK]

Introducing Gemini Decision 1, our new frontier decision model in the Gemini API.

It is $0.10 per 1M input tokens with no additional cost. With support for image, audio, video, and PDF natively, 256K context window, & 1K questions.

You can try it for free in the API!

[MirAI News @MirAI_Newz]

THIS IS HUGE!

[Rhys @RhysSullivan]

Executor now has first party OAuth clients for every major Google service

One click sign in and you're able to use as many accounts as you want with any agent

[MirAI News @MirAI_Newz]

MiniMax released H3-Image as open weights. It generates and edits images at about 16 megapixels natively, with no upscaler, and MiniMax says a 4K image takes 10 seconds.

It's a 33.1B diffusion transformer that runs in 6 steps, and it can edit with up to 16 reference images in one prompt.
https://t.co/JZSHzJukPo

[MirAI News @MirAI_Newz]

Google opens up Gemini 4 Carbon at Argon's price

Gemini 4 Carbon is now on the API for all paying developers, two weeks after Google gave it to a small group of cybersecurity defenders. Google calls it its strongest model, aimed at long multi-step coding, debugging and research.

• It costs $2 per million input tokens and $10 per million output, the same introductory price as Gemini 4 Argon. Carbon spends more tokens per task than Argon, so expect a bigger bill per task.
• Google says it ranks #1 on CursorBench 4.0, Perplexity's WANDR and Zapier's AutomationBench, at less than half the cost per task of the next best outside model.
• Google Antigravity gets a /swarm preview, where Carbon handles planning and auditing and lighter models do the parallel work.

https://t.co/A9BEDhMICG

[MirAI News @MirAI_Newz]

HERE WE GO BIG MODEL DROP!

[Francesco Oddo | AI Arc @oxfrancesco_]

Anthropic released Claude Fable 5.2 and Mythos 5.2, and cut Fable's price to $8 per million input tokens and $40 per million output, from $10 and $50.

The number I'd look at is FrontierCode 2.0. Fable 5.2 at High effort scores about the same as Opus 5.5 at Max, for roughly half the cost per task. At Xhigh it's the best released model Anthropic tested.

It's also up to 30% faster than 5.1, and a swarm of 1,000 Fable agents formalized Fermat's Last Theorem in 30 hours. Anthropic's own advice is still to start with Opus 5.5 for most coding, which is a modest pitch for a launch.

If you run Opus at Max on your hardest tasks, try Fable at High on the same tasks and compare the bill.

[MirAI News @MirAI_Newz]

Meta just shipped Muse Ember, its smallest model, priced at $0.10 per million input tokens and $0.50 per million output.

Alexandr Wang says it matches Claude Haiku 5.5's best score on the Artificial Analysis index for about a third of the cost per task. It also has 1M context and up to 512K output tokens.

On Meta's own benchmark table, Ember wins almost everywhere. The one coding test it clearly loses is FrontierCode, which checks whether the code is mergeable. That's a fair summary of what a subagent model is for.

The setup Meta suggests is Muse Wonder for planning and Ember for the volume work. If your agent bill is mostly file reads, test runs and small fixes, run a cost test this week.

[Alexandr Wang @alexandr_wang]

1/ announcing our smallest, most efficient little model is here: muse ember 🔥

built for work that agents repeat all day, it gets you more intelligence per dollar than Haiku, Luna, and all other flash models at any reasoning setting and does it at a third of the cost or less!

[MirAI News @MirAI_Newz]

OpenAI released Images 2.6: GIF generation with per-frame editing, saved reference images, and roughly 50% faster output. Live today in ChatGPT and Codex on every tier.

For the API there are now two models. GPT-Image-2.6 Flare does an image in about six seconds at low quality.

Sunburst is the slower, more consistent one for multi-round edits. Image markup moved from ChatGPT into the API, so you can mark the exact region to change, and SVG export is new.

https://t.co/meeIw7lxYW

[MirAI News @MirAI_Newz]

Hint:

[Chetaslua @chetaslua]

🚨 Not just Mythos 5.5. Fable 5.5 already hit Azure.

→ https://t.co/A4siM3A1Jd: "Mythos 5.5" appears 6× in the HTML (hidden h1 + logo alt text). The visible text still says 5.1

→ Oct 13: claude-fable-5-5 added to LiteLLM's Azure price list

→ Oct 14: removed, "Azure does not list"

→ MLflow's Azure model list still has it (1M ctx, $10/$50, the same as 5.1)

[MirAI News @MirAI_Newz]

This week is Astra and Fable week! Get ready boyz.

[MirAI News @MirAI_Newz]

Logan Kilpatrick announced Gemini 4 Carbon, which follows the Argon model released two weeks earlier.

This model looks BONKERS, that's what Argon should have been! I hope we get it soon, now it's only for Cyber Defenders.

https://t.co/iaogpyIZx8

[Logan Kilpatrick @OfficialLoganK]

Introducing Gemini 4 Carbon, it's a big jump over Argon which we released just 2 weeks ago.

Gemini 4 Carbon is SOTA on cyber, long horizon tasks, ranks #1 on Artificial Analysis V5, significantly improves on usability in domains like coding, and is more aligned.

[MirAI News @MirAI_Newz]

Mhhhhh let's see how it is.

[Harshith @HarshithLucky3]

GPT 6.1 Terra atleast be better than Sol in terms of speed

[MirAI News @MirAI_Newz]

Grok bot now has logins and cards... He can live for you basically.

[MirAI News @MirAI_Newz]

Go nuts Gemini Sommeliers

[Lumina @LuminaBench]

Gemini 4 Argon is now on Antigravity for Pro subs

Very quick rollout tbf, it has a separate limit now too

[Francesco Oddo | AI Arc @oxfrancesco_]

Happy to tell you that I've opened a bunch of slots on my calendar for a 30 mins free consultation about "How AI can improve your company!".

They can book here: https://t.co/Wn3AtrWglL

Refer a company and get a 15% success fee plus 1hr free consultation.

[MirAI News @MirAI_Newz]

SpaceXSI announced Grok Voice 3.0

Looks really good!

https://t.co/NquqS2eFHa

[SpaceX Super Intelligence @SpaceXSI]

Introducing Grok Voice 3.0, the world’s most intelligent speech-to-speech model.

It excels at resolving customer calls, responds in under a second, and costs less than other frontier models.

[MirAI News @MirAI_Newz]

WE HAVE IT GUYS! FINALLY! IT'S TESTING TIME!

[Logan Kilpatrick @OfficialLoganK]

Today we are rolling out Gemini 4 Argon globally in the Gemini API, to Pro + Ultra in Gemini App, and Ultra subs in Antigravity!

Argon is a huge step forward for Gemini, and a labor of love. This model has been battle tested by over 100,000 people at Google over the last month, with so much new innovation, ambition, and hard work baked into this release. The team has been heads down over the last 6 months working to make this possible across the stack (research, infra, safety, eng, etc).

We rolled the model out to cyber defenders as part of our fairwind program and early access customers over the last 2 weeks, it has been amazing to see and talk to folks who are now using this model daily. Seeing the models writing capability, raw knowledge and persistence for coding, paired with new features like 1M token output when the model needs to work more has gotten me even more excited about our path ahead.

This is just the start of the Gemini 4 era. As you saw with 3.5, 3.6, 3.7, and 3.8 Flash, our iteration speed with Gemini is fast : ) Now with Gemini 4 Argon to work from, I am cannot wait to see the progress continue, and most importantly see what this new model enables for the world.

This is such a special moment in the arc of technology, as millions of people everyday tackle problems they could never have thought possible before. Gemini 4 Argon was built to help you tackle the hardest problems. I have personally felt this spark with Argon and hope you do too. Onwards!!

[MirAI News @MirAI_Newz]

Gemini 4 Argon is here and Fable 5.5 is on the horizon. Is it gonna be like Gemini 3 Pro and Opus 4.5 release all over again?

You know how do they say, history doesn't repeat itself, but it rhymes?

[MirAI News @MirAI_Newz]

Alexandr Wang announced Muse Wonder! Their Fable class model. Let's see if it's benchmaxxed or really this goood.

https://t.co/zLphfFE9M3

[Alexandr Wang @alexandr_wang]

introducing muse wonder

our new frontier model. state-of-the-art at personal agents and professional agents.

our largest and most capable model yet—rivals the big bois

[MirAI News @MirAI_Newz]

JUICY

[Harshith @HarshithLucky3]

Muse Wonder 🍉🍉

[MirAI News @MirAI_Newz]

MAYBE today???? 😭

[MirAI News @MirAI_Newz]

Coming soooooon ™️