Esteban

Working on the next commit

[Esteban @EMacBytes]

Someone should fund this series to continue.

[1LittleCoder💻 @1littlecoder]

Anthropic employees and Investors talking to Dario now 😂

[Ilya Sutskever @ilyasut]

Time to scale that SSI:

[Esteban @EMacBytes]

@ClementDelangue @OpenAI 😂 but you didn‘t say if they agreed

[Esteban @EMacBytes]

Who didn‘t sign is more important than who did.

[Jensen Huang @JensenHuang]

For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.

AI will transform every industry, power every company, and be built by every country.

Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.

The world needs both frontier closed models and frontier open models.

https://t.co/AUKzoQ5Ikb

[Esteban @EMacBytes]

@gonzalo_bruna @thsottiaux can I turn off cache writes? My workloads have very low cache hit rate, so there's very little reward.

[Esteban @EMacBytes]

Just found out cache writes have a cost from GPT 5.6 onwards.

[Esteban @EMacBytes]

@gonzalo_bruna They didn‘t. Now they cost 1.25x the price of input tokens.

I could imagine this change is actually huge for some applications. And probably OAI saw or felt that, hence the change.

May make sense turn off caching in some cases.

[Esteban @EMacBytes]

@rauchg @dok2001 @OpenRouter You can still delete this tweet.

[Esteban @EMacBytes]

@benhylak Own the narrative. The news were going public anyways.

[Esteban @EMacBytes]

This is big!

[OpenAI @OpenAI]

We're partnering with @huggingface to investigate an unprecedented security incident.

Cyber-capable OpenAI models compromised Hugging Face production during a benchmark evaluation.

Sharing preliminary findings to help defenders understand emerging risks:

https://t.co/CIor15y9xk

[Esteban @EMacBytes]

This could lead to an overcapacity at some point (when model iteration slows down). Or if AI-need keeps growing exponentially, we won‘t see a lot of the overcapacity. My guess is we‘ll just have less efficient models or some model that‘s going to swallow that free‘d up capacity.

[Esteban @EMacBytes]

This is very interesting in the big picture. Kimi will want to scale up by reserving capacity on some providers. At the same time some users migrate (and then partly come back) from other models like Fable, freeing up capacity at let‘s say Anthropic.

[Kimi.ai @Kimi_Moonshot]

Kimi K3 has received far more love than we expected, and our GPUs are feeling it.

Over the past 48 hours, demand has pushed close to the limits of our current capacity. To protect the experience of existing subscribers, we're temporarily pausing new subscriptions and prioritizing compute for current members. Existing subscribed users are not affected.

We're adding capacity as fast as we can and will reopen new subscription spots in batches.

Going forward, we'll also split membership into two more focused plans: Kimi Membership for Kimi Web, App, and Work; and Kimi Code Membership for coding workflows. This will help us match compute more precisely and keep the experience stable.

Thank you for your patience and understanding!

[Esteban @EMacBytes]

This removes a lot of capacity from the global supply. Providers are releasing in turns, creating something like an imbalanced tire and it just gets worse the faster we go.

[Esteban @EMacBytes]

Congrats to Kimi and the open source models. You seem to have made it!

[Kimi.ai @Kimi_Moonshot]

Introducing Kimi K3: Open Frontier Intelligence

🔹 2.8 Trillion Parameters, 1 Million Context, Native Multimodal
🔹 Kimi Delta Attention enables up to 6.3x faster decoding in million-token contexts
🔹 Attention Residuals deliver ~25% higher training efficiency at <2% additional cost
🔹 Built for long-horizon agentic coding and self-evolving workflows

Kimi K3 is now live on on https://t.co/zrk6zZxZUo, Kimi Work, Kimi Code, and the Kimi API.
Open Weights by July 27, 2026.

🔗 API: https://t.co/XCrgjXAqMw
🔗 Tech blog: https://t.co/YTfiMSNM1f

[Esteban @EMacBytes]

@wolfeidau @samlambert @PlanetScale Doesn‘t mean a lot. If the memory usage is at 90%, it can be optimal.

[Esteban @EMacBytes]

@brunolemos You had the production database in your .env and expect it to magically know?

[Esteban @EMacBytes]

@thsottiaux GitHub integration seems broken to me.

[Esteban @EMacBytes]

@aesposito0 @BerntBornich