Nathan Del Cid

[Nathan Del Cid @nrdcid]

@yacineMTB me when i have the most minor runtime issue

[Thinking Machines @thinkymachines]

Today, we are introducing Inkling.

Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available.

https://t.co/Ghebq5mG30

Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵

[Nathan Del Cid @nrdcid]

@mattshumer_ That’s pretty wild that it called a full blown forced removal of whole user. In 2026.

We honestly need to scale interpretability tools at the same rate as these models

[Nathan Del Cid @nrdcid]

@pmddomingos Let him cook

[Nathan Del Cid @nrdcid]

@ArjChi I don’t think a singularity is fundamentally possible. That polarizing feeling is just the bubble getting ready to burst

[Elliot Arledge @elliotarledge]

Claude Fable 5 [max] wrote the first genuine (and fastest) megakernel ever submitted to KernelBench-Mega.

It was tested on: Kimi-Linear W4A16 batch-1 decode for RTX PRO 6000 Blackwell. Every prior model "won" it with a multi-kernel Triton pipeline that fails our single-fused-kernel authenticity gate

> Opus 4.8 at 14.4x
> GLM-5.2 11.1x
> GPT-5.5 4.3x
> Sonnet 5 4.0x.

Fable shipped 18.7x over reference, and torch.profiler shows exactly ONE cooperative kernel launch per decoded token. Int4 dequant (nibbles unpacked in-register, never materialized), conv+SiLU, KDA gated-delta state, MLA absorbed-latent attention with online softmax, MoE router + top-8 experts, RMSNorms, even the KV cache append all inside one launch, staged by 14 grid barriers. We overwrote its input buffers mid-audit to prove it recomputes on live data. It does.

The advantage grows with context. 17.8x at 2k, 18.9x at 8k, 19.5x at 16k. Longer context means a bigger KV cache and more attention work per token which is usually where a decode kernel bleeds. Keeping everything in one launch amortizes the fixed barrier overhead and the int4 GEMV stays bandwidth-bound, so the gap over the reference widens instead of closing.

It spent 64% of the session in silence timing the baseline, microbenchmarking grid barriers, deriving a ~29x bytes/token roofline, then wrote the whole kernel once, hit 14.4x on the first benchmark, and spent the last hour deleting barriers and making int4 dequant free (one LOP3 + HSUB2/HMUL2). The one regression it tried (finer split-K) it measured and reverted instead of rationalizing.

https://t.co/eXKN2sZMVr

[Nathan Del Cid @nrdcid]

@beffjezos @midjourney X @DecartAI

[Nathan Del Cid @nrdcid]

@sun_hanchi 1 is not controversial it’s actually just facts because the bottleneck is compute which mainly requires EE/physicists

[Nathan Del Cid @nrdcid]

@angelizado2009 @TIDAL It's not a stupid question because the post clearly states that "100% AI-generated songs will receive a badge", so I'm wondering if that would still be the case for slightly < 100% and how they plan to measure that criteria :)

[Nathan Del Cid @nrdcid]

@angelizado2009 @TIDAL I don't make music at all, AI included. It was just a question :P

[Nathan Del Cid @nrdcid]

@_yizzuz_ @NousResearch @Shpigford @lmstudio

[Nathan Del Cid @nrdcid]

@dwarkesh_sp It’d be better off to just integrate LLMs at the kernel level than to have it run on the application layer

i.e. we need OS renovations

[Nathan Del Cid @nrdcid]

@jxnlco Does Computer Use utilize Microsoft’s Omniparser?

[Nathan Del Cid @nrdcid]

@elonmusk @CoreyTheX @XMoney I can truly and with 100% certainty say that I have never, EVER, in my life seen anyone abbreviate “thanks” like that until now

[Nathan Del Cid @nrdcid]

@trikcode Open up any AI textbook from the 90s and you’ll see why.

“Agents” are a fundamental concept in artificial intelligence, and LLMs are more or less “new” but surprisingly effective (we literally don’t know why), so we’re trying to fit them into the definition of an agent.

[Nathan Del Cid @nrdcid]

@gabriel1 This is what the government is scared of

[François Chollet @fchollet]

The true measure of a software engineer isn't their ability to write clever code. It's their ability to ruthlessly protect the codebase from unnecessary cleverness.

[Nathan Del Cid @nrdcid]

@paulg I'm of the opinion that the compiler is just now starting to recognize this. There is a certain fatigue a-brewing

[Aleph @alephneuro]

We recently obtained the highest-resolution 3D images of the human brain ever taken from outside the skull. This is the first look.

Introducing Aleph, a research lab building brain interfaces for the telepathic future. (1/n)