顶尖科技推文 - 2026年7月2日

2026年7月2日科技每日简报

Today's top tech conversations are led by @alex, whose post about 'RT @AnthropicAI: Claude Fable ...' garnered the highest engagement. Key themes trending across the top stories include sonnet, claude, effort, model, compute. The community is actively discussing recent developments in AI, engineering practices, and startup strategies.

1. alex (Group Score: 1378.4 | Individual: 65.9)

Cluster: 37 tweets | Engagement: 12404 (Avg: 369) | Type: Tech

RT @AnthropicAI: Claude Fable 5 will be available again globally tomorrow.

After a series of productive conversations with the US government, we're redeploying the model with a new set of classifiers to target and block more cybersecurity tasks. In the near term, some routine tasks like coding and debugging will fall back to Opus 4.8. We’ll continue to refine these classifiers over the coming weeks to reduce false positives and better distinguish genuine misuse from legitimate requests.

We’ve also begun drafting a consensus framework—with Amazon, Microsoft, Google, and other Glasswing partners—for assessing the severity of AI jailbreaks and how AI developers should respond to them. We invite other industry partners and model providers to join us in this effort.

Finally, we’re scaling up our collaboration with the US government on model testing and safeguards. This will include pre-release access to models and safeguards for evaluation, information sharing on jailbreaks and misuse, and dedicated resources for joint research.

Thank you to our users for your patience, and to our partners across the government, industry, and the research community who worked alongside us to make Fable 5 available again.

Read our full blog: https://t.co/VHyum831ri

See 36 related tweets

@tunguz: Whelp, no Fable for coding any time soon after all. :/\n\nQT @AnthropicAI: Claude Fable 5 will be av...
@Prince_Canuma: Local AI must win!

They nerfed Fable 5 even further 🤔 https://t.co/X0XcxleR14\n\nQT @AnthropicAI: C...

@minchoi: Fable 5 is back...

what> https://t.co/SfWvehnGVE\n\nQT @AnthropicAI: Claude Fable 5 will be avai...

@AlexFinn: Fable 5 will be API pricing starting July 7th.

For the next week, you can use up to 50% of your usa...

@banteg: > a new set of classifiers to target and block more cybersecurity tasks

> routine tasks like ...

2. MadisonMills22 (Group Score: 558.2 | Individual: 45.4)

Cluster: 32 tweets | Engagement: 25812 (Avg: 1302) | Type: Tech

RT @AnthropicAI: We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5.

We'll begin restoring access tomorrow, and will share an update soon.

We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models.

See 31 related tweets

@Teknium: Hopefully this doesn’t happen again. Excited to see what gpt 5.6 Sol + Fable produces with our MoA!...
@teortaxesTex: that's good some people were starting to develop the illusion that Fable was never all that. It'll s...
@SophiaCai99: Commerce has alerted Anthropic, which says it will begin restoring access tomorrow.\n\nQT @Anthropic...
@noahzweben: A dream come true A real life fairytale A ... Fable\n\nQT @AnthropicAI: We’ve received notice that t...
@CNBC: Anthropic on Tuesday said the U.S. Department of Commerce has lifted the export controls on its Clau...

3. wallstengine (Group Score: 444.1 | Individual: 41.6)

Cluster: 16 tweets | Engagement: 679 (Avg: 111) | Type: Tech

$META is reportedly developing a cloud business to sell access to excess AI compute, per Bloomberg.

The internal initiative is called Meta Compute.

The plans being considered:

AI model access hosted on Meta infrastructure, similar to AWS Bedrock

Raw AI compute capacity, closer to CoreWeave

Developer access to Meta’s data centers, chips, and models

See 15 related tweets

@ChrisRMcGuire: And then there were three.

Non-frontier AI labs (first xAI, now Meta) are pivoting away from AI de...

@MilkRoadAI: RT @MilkRoadAI: Meta is building a cloud business to sell excess AI computing capacity to outside cu...
@StockMKTNewz: Meta Platforms $META stock is now up by more than 7% in pre market trading

🟢🟢🟢🟢🟢🟢🟢 https://t.co/XB...

@WOLF_Financial: META $META IS BUILDING A CLOUD BUSINESS TO SELL AI COMPUTE TO OUTSIDE CUSTOMERS

Meta stock jumped 8...

@BullTheoryio: BREAKING: $META is now up 10.78%, adding roughly$ 170 billion in market value today.

Meta's massive...

4. zephyr_z9 (Group Score: 433.5 | Individual: 66.3)

Cluster: 13 tweets | Engagement: 1983 (Avg: 172) | Type: Tech

LMAO Zuck finally takes the L\n\nQT @wallstengine: $META IS BUILDING A CLOUD BUSINESS TO SELL EXCESS AI COMPUTE https://t.co/ZIwove3ljD

See 12 related tweets

@wallstengine: $META IS BUILDING A CLOUD BUSINESS TO SELL EXCESS AI COMPUTE https://t.co/ZIwove3ljD...
@StockMKTNewz: JUST IN:

META PLATFORMS $META IS BUILDING A CLOUD BUSINESS TO SELL EXCESS AI COMPUTE https://t.co/P...

@jukan05: Meta falls out of the AI race.\n\nQT @wallstengine: $META IS BUILDING A CLOUD BUSINESS TO SELL EXCES...
@tunguz: If you fail as a top AI lab, you can always become another neocloud.\n\nQT @amitisinvesting: $META

...

@scaling01: now that was predictable\n\nQT @wallstengine: $META IS BUILDING A CLOUD BUSINESS TO SELL EXCESS AI C...

5. MTSlive (Group Score: 403.4 | Individual: 48.6)

Cluster: 20 tweets | Engagement: 1767 (Avg: 173) | Type: Tech

me and fable 5 tomorrow morning https://t.co/SMbMfm2tMu\n\nQT @MTSlive: SITUATION UPDATE: Access to Claude Fable 5 will begin rolling out to users tomorrow.

See 19 related tweets

@shiri_shh: developers when Fable 5 is back in Claude again tomorrow. https://t.co/0xMOENeWm8\n\nQT @AnthropicAI...
@kimmonismus: Fable 5 is back, but with a major caveat.

Coding is being handled even more restrictively and route...

@alexcooldev: Wake up, guys. Claude Fable 5 and Mythos 5 are coming back tomorrow. I’m about to resubscribe to my ...
@trq212: excited to get Fable back in your hands!\n\nQT @AnthropicAI: We’ve received notice that the Departme...
@synthwavedd: "claude-fable-5-v2" was briefly added to @arena in Battle Mode but has now been removed (probably pr...

6. bubbleboi (Group Score: 368.1 | Individual: 56.2)

Cluster: 12 tweets | Engagement: 3714 (Avg: 161) | Type: Tech

RT @IntCyberDigest: ‼️ BREAKING: Anthropic has embedded hidden spyware-like code in Claude Code that covertly targets Chinese users. It then sends information regarding every user by injecting it into their prompt message.

Claude Code is sending info like timezone, proxy and possible AI Lab connections into the system prompt in ways Chinese users can't notice.

A coding agent with repo and command permissions should not silently hide routing metadata inside prompts. This is a serious breach of user trust.

See 11 related tweets

@Teknium: This is pretty concerning. You could still do this at the API level to some degree, but they seeming...
@0xSero: Can it get any worse 😭\n\nQT @IntCyberDigest: ‼️ BREAKING: Anthropic has embedded hidden spyware-lik...
@zephyr_z9: LMAO\n\nQT @IntCyberDigest: ‼️ BREAKING: Anthropic has embedded hidden spyware-like code in Claude C...
@elder_plinius: new prompt injection technique just dropped\n\nQT @IntCyberDigest: ‼️ BREAKING: Anthropic has embedd...
@Hesamation: Anthropic confirmed this.

“Claude Code is sending info like timezone, proxy and possible AI Lab co...

7. StockSavvyShay (Group Score: 355.7 | Individual: 43.3)

Cluster: 12 tweets | Engagement: 1597 (Avg: 594) | Type: Tech

Neo-cloud names are moving lower after reports that $META is developing “Meta Compute” a cloud business to sell excess AI compute.

The market is reading this as potential new hyperscaler supply in AI compute which could pressure standalone GPU cloud pricing.

• $NBIS -10% •$ CRWV -10% • $IREN -5% •$ CIFR -5% • $WULF -5% •$ APLD -4%\n\nQT @StockSavvyShay: $META IS BUILDING A CLOUD BUSINESS TO SELL EXCESS AI COMPUTE https://t.co/zdeT7IY1Wf

See 11 related tweets

@zerohedge: *META IS BUILDING A CLOUD BUSINESS TO SELL EXCESS AI COMPUTE

First SpaceX, now Meta selling somethi...

@StockSavvyShay: $META is up over 7% after announcing plans to build a cloud business to sell excess AI compute. http...
@BullTheoryio: BREAKING: $META is up 8.6% in premarket after Bloomberg reported the company is building a cloud bus...
@rohanpaul_ai: Mandeep Singh from Bloomberg on Meta's move to cloud computing Meta’s $150B CapEx now needs ROI; re...
@coinbureau: 🚨ALERT: META SURGES 10% ON AI CLOUD PUSH

Meta shares surged over 10% after Bloomberg reported the c...

8. Meer_AIIT (Group Score: 208.3 | Individual: 32.5)

Cluster: 8 tweets | Engagement: 18 (Avg: 107) | Type: Tech

most "legal ai" just summarizes a pdf and calls it a day.

ivo benchmarks does the thing a wrapper can't.

it pulls your team's entire negotiating history into the contract you're reviewing.

then scores every clause against how you've actually handled it before.

→ how often you've agreed to that provision → whether to accept or push back → a link to the deals where you held the line

your negotiating history used to live in one senior attorney's head.

or buried in a clm no one trusts.

or lost in a shared drive.

now it shows up in every review, on every clause.

every other business function already got its system of intelligence.\n\nQT @mkjung_: Introducing Ivo Benchmarks.

Benchmarks reviews and redlines agreements by drawing upon your company's entire history of contract negotiations.

We developed Benchmarks by working closely with some of the world's leading companies. They have literally millions of contracts, but no feedback loop for improving their negotiations based on what they've agreed to in the past.

What they need is institutional memory for their negotiations. We're excited to support that by making Benchmarks generally available.

The Ivo team is working on a lot of really cool things right now. Much more coming soon!

See 7 related tweets

@rohanpaul_ai: This is truly a great use of AI.

Ivo Benchmarks is turning a company’s legal contract review histo...

@VadimStrizheus: Do you understand what just happened?!

IvoAI just launched the OpenClaw but for legal teams.

It ...

@Mayhem4Markets: Every contract your legal team has ever negotiated is sitting in a drawer.

Not metaphorically. In a...

@cgtwts: Every company has years of negotiation experience, but it’s often hidden in old contracts or remembe...
@shiri_shh: Lawyers have been feeding confidential contracts into ChatGPT for years.

Nobody in legal would ever...

9. WesRoth (Group Score: 173.9 | Individual: 31.3)

Cluster: 7 tweets | Engagement: 60 (Avg: 19) | Type: Tech

Sonnet 5 is cheaper per token than Opus, but not necessarily cheaper per completed job.

The model reaches stronger results by working longer, producing more tokens, and taking more agentic steps.

For developers, the real metric is no longer the price per million tokens. It is the total cost of getting a reliable result.\n\nQT @ArtificialAnlys: Claude Sonnet 5 achieves 53 on the Artificial Analysis Intelligence Index, but without promotional pricing will cost more per task than Opus 4.8

We supported @AnthropicAI to evaluate Claude Sonnet 5 ahead of release: with max effort it improves 6 points over Sonnet 4.6 to achieve the same Intelligence Index as GPT-5.5 with high reasoning, but remains behind Opus 4.7 and 4.8

Key takeaways:

➤ Claude Sonnet 5 is the #5 model on the Artificial Analysis Intelligence Index, only 2-3 points behind GPT-5.5 (xhigh) and Opus 4.8 (max)

➤ With max effort, Sonnet 5 works harder than previous Anthropic models: it used ~40% more output tokens per Intelligence Index task than Sonnet 4.6, and ~3x the agentic turns for our knowledge work evaluations AA-Briefcase and GDPval-AA. This behavior scales well with the ‘effort’ setting, with the max effort using around 6x more turns than low effort on GDPval-AA

➤ Claude Sonnet 5 costs more per task than Opus 4.8 before accounting for promotional pricing: Claude Sonnet 5 costs $2.29 per task on the Intelligence Index, a ~2x increase compared to Sonnet 4.6 and ~15% more than Claude Opus 4.8. This is driven entirely by increased token usage. Sonnet 5 retains the same$ 3/ $15 per 1M input/output token pricing as Sonnet 4.6 (compared to$ 5/ $25 for Opus 4.8), however Anthropic is offering a one-third reduction to$ 2/ $10 until September 1. Our results use standard$ 3/$15 pricing

➤ Sonnet 5 matches or outperforms Opus 4.8 on agentic knowledge work tasks: on both AA-Briefcase and GDPval-AA, Claude Sonnet 5 sits just ahead of Opus 4.8, trailing only Claude Fable 5 (which is not currently generally available). These benchmarks test the ability of models to produce accurate and well-presented professional outputs using our open source reference agent harness, Stirrup

➤ For reasoning and knowledge-heavy tasks, Sonnet still sits behind its larger siblings: despite substantial gains across many evaluations, heavy reasoning and knowledge benchmarks still show Opus 4.8 ahead of Sonnet 5. On CritPt, a frontier physics reasoning benchmark developed by researchers at Argonne and UIUC, Sonnet 5 scores 17% - this is 14 points higher than its predecessor, but behind GLM-5.2, Claude Opus and Fable, and GPT-5.5 (xhigh and Pro)

➤ Sonnet 5 also showed significant improvements over Sonnet 4.6 on Terminal-Bench v2.1 (+9 points), Humanity’s Last Exam (+10 points), and SciCode (+7 points), with relatively flat scores elsewhere

Other key model details:

➤ Context window of 1 million tokens (equivalent to Sonnet 4.6)

➤ Pricing of $3/$ 15 per 1M tokens of input/output (reduced to $2/$ 10 until September 1); cache pricing remains at a 25% premium for cache writes ( $3.75 per million tokens) with 5-minute time to live, and 90% discount for cache hits ($ 0.3 per million tokens)

➤ Effort remains the recommended way of configuring model performance and latency. Sonnet 5 adds an additional ‘xhigh’ effort setting relative to Sonnet 4.6, matching the 5 effort levels available on Opus 4.8 (max, xhigh, high, medium, low)

See 6 related tweets

@edzitron: This shit is so funny to release after GLM 5.2. Anthropic truly thinks it has everybody under its sp...
@MTSlive: SITUATION ANALYSIS: Claude Sonnet 5

At long last, a new 5-series model from Anthropic! Is Fable 5 b...

@TeksEdge: This is a big deal Sonnet 5 has scored a 53 putting in a GPT-5.5-high class model available to anyon...
@TeksEdge: GLM-5.2 is the smartest open-source model available right now, but you'll pay a premium for its outp...
@Marktechpost: RT @Marktechpost: Most "cheaper model" launches just trade quality for price. Claude Sonnet 5 is doi...

10. genspark_ai (Group Score: 156.1 | Individual: 31.6)

Cluster: 7 tweets | Engagement: 43 (Avg: 36) | Type: Tech

🚀Claude Sonnet 5 is now live in Genspark AI Chat Agent, Code Agent and Claw.

Anthropic’s most agentic Sonnet yet — it plans, uses tools, and runs autonomously at a level that once required much larger models.

Try it now! → https://t.co/TLYh9dJlXf\n\nQT @claudeai: Introducing Claude Sonnet 5, our most agentic Sonnet yet.

It makes plans, uses tools like browsers and terminals, and runs autonomously at a level that just a few months ago required larger and more expensive models. https://t.co/UKK8G7ww5h

See 6 related tweets

@minchoi: Claude Sonnet 5 is here.

Not the cheapest on every benchmark.

But agentic Claude just became the d...

@Gorden_Sun: Claude发布Claude Sonnet 5 初步感受是代码能力一般，但是写作和思考能力很不错，可以考虑在写作任务里替换掉Opus 4.6\n\nQT @claudeai: Introducing ...
@coinbureau: 🚨NEW: ANTHROPIC UNVEILS CLAUDE SONNET 5

Claude Sonnet 5 is said to deliver near-Opus 4.8 performan...

@moneycontrolcom: #TechWithMC | Anthropic launches Claude Sonnet 5 with improved coding, reasoning and lower API prici...
@Cointelegraph: 🔥 UPDATE: Anthropic introduced Claude Sonnet 5, calling it its most agentic Sonnet model yet. https:...

11. Dan_Jeffries1 (Group Score: 151.9 | Individual: 48.4)

Cluster: 4 tweets | Engagement: 1730 (Avg: 291) | Type: Tech

RT @PalantirTech: Our thoughts on the importance of AI sovereignty.

Your AI sovereignty dictates your institution’s future. Sovereignty is the precondition for choice. Relinquishing sovereignty transfers the future choices of your institution to others, who are likely to exploit it for their gain and your loss.
Data retention is your treasure. Transfer it at your own peril. Your ability to win is dictated by your ability to recognize and use your unique edges, and you keep winning by compounding the underlying data to generate new insights. Transferring that data hands over access to your pre-existing winning plays and yields the means of production for new ones.
Tokenmaxxing hijacks your value orientation and decreases your institutional fortitude and intelligence. The pursuit of high token usage incentivizes disposable scripts over robust software — with the addictive feeling of false progress. There is a reason why those selling tokens refuse to charge based on value.
Controlling your weights is controlling your fate. Weights are the distilled form of hard-won, accumulated institutional knowledge. If you let others control your weights, you are allowing them to migrate the alpha of your business to theirs.
There is no contradiction between sovereignty and alpha. The architecture that maximally preserves sovereignty is one that enables institutions to own their tribal knowledge, and to compound it as alpha.
Politicizing the technical issues involving sovereignty is what your adversary wants. Techno-politicization is the wellspring of false sovereignty. Techno-politicization drives decisions that seem to reduce dependency, but ultimately limit agency — especially on the battlefield in the West.
Real expertise is existential. Allowing politics or favoritism to determine your technical decisions rewards whoever is best at politics, not whoever is right. Listen to those closest to the problems, not those speaking most compellingly about them.
Learn from institutions that are winning or that have consistently delivered. Institutions facing existential threats do not have the luxury of making technical decisions based on political preferences.
Only listen to institutions, countries, and people who have a proven record of being right. A track record of correctness is the best and only signal for future correctness. Judging something as right or wrong based on who you like is exceedingly misguided.

See 3 related tweets

@BrianRoemmele: Palantir is full on manifesto for opne source.

They are one of us now.

Join us.

AI sovereignty fo...

@Hesamation: Controlling your weights is controlling your fate. If you let others control your weights, you are a...
@JordanSchachtel: Nailed it\n\nQT @PalantirTech: Our thoughts on the importance of AI sovereignty.

Your AI soverei...

12. Scobleizer (Group Score: 149.3 | Individual: 36.4)

Cluster: 5 tweets | Engagement: 708 (Avg: 169) | Type: Tech

RT @xai: Introducing Voice Agent Builder: a no-code platform to create human-like voice agents with Grok Voice.

Available today at $0.05 / min.

https://t.co/kUkF7zqvfR https://t.co/OCIq1oDYar

See 4 related tweets

@testingcatalog: xAI announced a new Voice Agent Builder available on xAI Console.

Users can use the conversational...

@XFreeze: Creating AI voice agents has become ridiculously easy

xAI’s Grok Voice Agent Builder lets you deplo...

@Baconbrix: everyday we get closer to jarvis-as-a-service\n\nQT @xai: Introducing Voice Agent Builder: a no-code...
@teslaownersSV: BREAKING: xAI has introduced Voice Agent Builder.

The new no-code platform lets developers create h...

13. ivanfioravanti (Group Score: 142.9 | Individual: 51.6)

Cluster: 4 tweets | Engagement: 1564 (Avg: 181) | Type: Tech

RT @Etched: We're coming out of stealth.

We've built our first racks after a successful A0 tapeout, $1B+ in customer contracts, and$ 800m raised.

Early customer tests show us achieving SOTA throughput, latency, and power efficiency on inference workloads.

Our first racks ship this summer. https://t.co/FLccrkLTza

See 3 related tweets

@WesRoth: AI chip startup Etched has come out of stealth after producing its first inference racks and complet...
@ArtificialAnlys: Congratulations to @Etched on announcing their first inference system!

We have seen early performa...

@PatrickMoorhead: It’s lazy to refer to @Etched as a Theranos with “$1B+ in customer contracts” and the experience of ...

14. deanwball (Group Score: 141.0 | Individual: 60.7)

Cluster: 5 tweets | Engagement: 827 (Avg: 87) | Type: Tech

Great news! But we have no idea what Anthropic did to make the models “safe,” what commitments Anthropic has made going forward, and whether or how any of this applies to other frontier models in the government’s licensing queue. We know that GPT 5.6 is in that queue, but it’s fair to assume that other model developers are at least in early stages of submitting their models.

This opacity will not lend itself well to a stable, investable, trustworthy industry over time. But USG needn’t figure all this out in a day, and a two-week review timeline is not insane in the grand scheme of things. The status quo is not tenable, but we made progress today: the Anthropic models are available worldwide. That progress is just a first step, but it is worthy of applause nonetheless.\n\nQT @AnthropicAI: We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5.

We'll begin restoring access tomorrow, and will share an update soon.

We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models.

See 4 related tweets

@pstAsiatech: Yeah, now GLM-5.2 also in the queue..\n\nQT @deanwball: Great news! But we have no idea what Anthrop...
@ShanuMathew93: We’re so back

But in seriousness, we are in now an area where governments have direct access and in...

@firstadopter: Finally. U.S. government needs to get its act together.\n\nQT @AnthropicAI: We’ve received notice th...
@VaibhavSisinty: The US Department of Commerce just lifted export controls on Claude Fable 5 and Mythos 5.

Access to...

15. 0x_kaize (Group Score: 137.8 | Individual: 33.4)

Cluster: 5 tweets | Engagement: 37 (Avg: 52) | Type: Tech

the old company needed the founder online to keep momentum

the new company has agents running the loop while the founder is away

humans go offsite

agents build

humans take breaks

agents test

humans log off

agents fix

the company does not pause

that’s the actual Polsia demo\n\nQT @Bencera: I raised $30M for my zero employee startup.

My AI told me we should do an offsite to align on strategy.

So I rented a mansion and a yacht in Cabo and flew the whole team out.

Well, it was only me… and Polsia.

Presenting episode 2 of aisloP, a docu-series on how I build Polsia:

“The Offsite”

See 4 related tweets

@AnatoliKopadze: This might be the most triggering founder video of the year.

Trap music, vacation clips, a guy who ...

@shiri_shh: Bro is out here posting holiday clips while his startup builds itself

The new flex isn't looking bu...

@VadimStrizheus: pov: you've been working a 9-9-6 for the past 10 years, and this guy just raised $30M for an AI slop...
@WhaleInsider: JUST IN: Smart founders are optimizing for time away, not time worked.\n\nQT @Bencera: I raised $30M...

16. zerohedge (Group Score: 132.2 | Individual: 41.9)

Cluster: 5 tweets | Engagement: 2317 (Avg: 626) | Type: Tech

META will be the first to cut capex\n\nQT @zerohedge: *META IS BUILDING A CLOUD BUSINESS TO SELL EXCESS AI COMPUTE

First SpaceX, now Meta selling something called "excess compute"

See 4 related tweets

@benitoz: Meta exploring an AI cloud business is not the same as Meta admitting it overbuilt

The report is th...

@edzitron: yeah the argument they're going to make is that the neocloud thing Meta is doing will "fund future c...
@StockSavvyShay: $META up more than than 11% today on reports its developing a cloud infrastructure business to sell ...
@zerohedge: *META SHARES OPEN 7.9% HIGHER, MOST SINCE APRIL 8\n\nQT @zerohedge: META will be the first to cut ca...

17. zephyr_z9 (Group Score: 130.4 | Individual: 36.7)

Cluster: 6 tweets | Engagement: 288 (Avg: 172) | Type: Tech

Even the auto guys are signing LTAs\n\nQT @MicronCEO: Memory and storage are in the tightest supply the industry has seen in years. Critical industries like automotive, medical and defense need certainty as much as they need capacity.

Today Micron and General Motors expanded our strategic partnership. Through this strategic customer agreement, GM secures long-term supply of LPDRAM, NOR and UFS NAND, the memory and storage behind AI-enabled in-cabin experiences and advanced driver assistance in next-generation vehicles.

This is backed by capital already deployed, not just intent. Our $2 billion modernization of the Manassas, Virginia fab began production this year, giving GM the U.S.-based, long-cycle supply that automotive platforms are built on.

Mary Barra said it well: “Delivering next-generation vehicles at scale requires a resilient and closely aligned supply chain. Our expanded collaboration with Micron strengthens our access to critical memory technologies.”

We’re grateful for GM’s trust in Micron over decades. This is what two great American companies deliver when they commit to each other for the long term, supply predictability in a market that needs it, and a technology roadmap that keeps vehicles safer, smarter and more capable.

Proud of what two teams are building together.

https://t.co/Ak8EoEPYmU

#Semiconductors #Automotive #Manufacturing

See 5 related tweets

@wallstengine: Micron $MU and$ GM signed a strategic customer agreement to secure long-term supply of Micron memory...
@StockMKTNewz: Micron $MU and General Motors$ GM announced a Strategic Customer Agreement

"to secure a long-term, ...

@StockSavvyShay: $MU and$ GM signed a strategic customer agreement to secure long-term Micron memory and storage supp...
@negligible_cap: *MICRON & GENERAL MOTORS SIGN PACT TO SECURE SUPPLY

"Micron and GM continue to collaborate on f...

@pstAsiatech: Micron, GM sign have signed a long-term supply agreement for memory and storage ‌platforms used in ...

18. WesRoth (Group Score: 128.7 | Individual: 29.7)

Cluster: 6 tweets | Engagement: 38 (Avg: 19) | Type: Tech

Google is attacking the biggest barriers in generative media: waiting time and cost.

Google introduced Nano Banana 2 Lite and Gemini Omni Flash for the Gemini API and Google AI Studio.

Nano Banana 2 Lite is built for fast, low-cost image generation. It produces a 1K image in under four seconds at roughly $0.034, while still scoring competitively against larger image models.

It reached an Elo score of 1251 for image generation and 1308 for image editing, placing it close to Nano Banana 2 while running five times faster.

Gemini Omni Flash focuses on high-quality video generation and conversational editing. It delivers state-of-the-art video-editing performance at $0.10 per second, matching the price of Veo 3.1 Fast.\n\nQT @OfficialLoganK: Introducing Nano Banana 2 Lite 🍌 and Gemini Omni Flash 🔮, our new generative media models in the Gemini API and AI Studio!

Nano Banana 2 Lite is extremely fast (<4s image) & cheap ($0.034 / 1K image).

Omni Flash is SOTA at video editing at $0.10 / sec, same as Veo 3.1 Fast! https://t.co/qDxRpqpX5E

See 5 related tweets

@WesRoth: Google released two new creative AI models: Nano Banana 2 Lite and Gemini Omni Flash.

Nano Banana 2...

@moneycontrolcom: #TechWithMC | Google launches Nano Banana 2 Lite and Gemini Omni Flash for faster AI image and video...
@drivascos: RT @nbrichtova: Few updates today! We shipped:

Nano Banana 2 Lite: It's cheaper, faster, and bett...

@rohanpaul_ai: RT @rohanpaul_ai: Google released Nano Banana 2 Lite, a 4-second image model, alongside Gemini Omni ...
@Saboo_Shubham_: RT @Saboo_Shubham_: The MATH here is INSANE for Multimodal AI agents.

Nano Banana 2 Lite: ~3 cents ...

19. jason (Group Score: 124.8 | Individual: 35.7)

Cluster: 4 tweets | Engagement: 161 (Avg: 285) | Type: Tech

NotebookLM is one of the best products from Google in a long time

10x the funding on this @sundarpichai and let them keep cooking!

Don’t Google reader them!\n\nQT @NotebookLM: Doom scrolling but make it educational 🤓

Introducing Short Video Overviews in NotebookLM! Turn your most complex sources into 60-second, vertical videos that deep dive into any concept.

Rolling out now to Google AI Ultra and Pro subscribers on mobile & web (free users soon!) https://t.co/vaOa4ELbw4

See 3 related tweets

@WesRoth: NotebookLM is introducing Short Video Overviews, a new feature that turns complex source material in...
@chrisalbon: Google is cooking with NotebookLM. Here is an example: https://t.co/VaM7IVojUz\n\nQT @NotebookLM: Do...
@joshwoodward: RT @NotebookLM: Doom scrolling but make it educational 🤓

Introducing Short Video Overviews in Noteb...

20. firstadopter (Group Score: 124.2 | Individual: 36.0)

Cluster: 4 tweets | Engagement: 173 (Avg: 87) | Type: Tech

Ok, since this cryptic FUD tweet is hitting stocks (with an assist from @citrini ). I'll start:

What is Andrew talking about?
Who is Andrew?

Thanks.\n\nQT @AndrewCurran_: I'm posting this prediction now so I can quote it later. There has been a significant breakthrough in architecture - specifically around memory efficiency - not by one of the big labs, but by a team that was spun out of OpenAI (not SSI). They will probably announce it soon.

See 3 related tweets

@teortaxesTex: weirdly specific\n\nQT @AndrewCurran_: I'm posting this prediction now so I can quote it later. Ther...
@kimmonismus: If true, this would be much bigger than just another model release.

Memory efficiency is one of th...

@IamEmily2050: Jerry team 👁️👁️👁️\n\nQT @AndrewCurran_: I'm posting this prediction now so I can quote it later. The...