jefflau.eth (@Jefflau) header
jefflau.eth (@Jefflau) profile picture

Co-founder @ensdomains. Distilling the complicated into the simple. In local AI psychosis.

Joined December 2010
Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)Photo by jefflau.eth (@Jefflau)
Thread of my threads distilling the complicated into the simple. Previous topics include: - Zero Knowledge Proofs - EIP-1559 - Rollups - ETH2 and the merge I’ll add new threads here for easy referencing for myself and others.
42
97
5
558
This is pretty much exactly why i’m bought in on local AI. Even if intelligence stopped today, i’d be happy with what my local agents can run for 99% of things and likely things are only going to get better. Also there’s a good point about customer data. There’s way more to go wrong when you have to share that with third parties. Same thing with your own personal or family data.
I was asked a few times this past week why I bought a super expensive machine for local inference. Why not rent? One side of the answer is that I want to automate some things for Dark Bio, like accounting and invoicing. I refuse to send customer data to 3rd party providers. So that's a practical necessity I can more easily rationalize. The other side, however, is a hedge against LLM regulations. Restricting access to model weights themselves is impossible; software piracy existed forever and one more law against file sharing won't make a dent in curbing that. That said, restricting access to hardware is a *very* effective way to regulate things. These machines are not cheap things that fly under the radar; they trigger red flags across all payment and shipping flows, so one single law passed and all flows stop instantaneously. And already today, you need to fill out forms to be allowed to buy a DGX Spark. My concern is that with LLMs advancing and various hacks happening, some regulator will make the unfortunate knee-jerk reaction to block hardware sales. And even if it takes weeks or months to pass, it might be too late to obtain the required local inference capability with the current 3-4 month entry queues. Is this paranoia? Probably. Is this a waste of money? Possibly. But potentially having access cut off to both hosted frontier models and local inference overnight is a pretty nasty setback with no meaningful way to paddle out of.
8
13
1,052
Gpus for everyone!
We're giving away a Hogwarts Legacy custom wrapped NVIDIA GeForce RTX 5090 to celebrate its official release with DLSS 5 🪄 To enter: 🟢 Share this post 🟢 Comment #RTXON Terms and Conditions: nvidia.com/en-us/geforce/con…
3
9
597
With the recent wave of vulnerability disclosures, it’s a good time to take a closer look at your own code. Turn your coding agent into a team of security auditors with the open-source skill that seeded Cloudflare’s vulnerability discovery harness. The 6 phase approach: 1. Map the codebase 2. Hunt for vulnerabilities 3. Challenge each candidate 4. Record the findings 5. Verify the findings 6. Report what’s confirmed, rejected, or still needs validation Start your hunt here: github.com/cloudflare/securi…
236
1,584
293
18,167
2,446,863
Soon. 4x RTX PRO 6000. I heard Sam said we shouldn’t own our RAM. Apparently he doesn’t want us to own our own intelligence 🤷‍♂️
GLM 5.3 Flash on 2x RTX PRO 6000s
5
14
787
This is scary
Yesterday, an anonymous black hat group reached out with an insane RL request. They offered an astronomical amount of money to post-train open-weight models on exploit data. The amount of attack data they claim to hold is unlike anything I have ever seen. If someone told me this was an undercover FBI dataset, I would have believed it. For context, a white-hat project I worked on using GLM-5.3 used less than one percent of that volume, and it already outperformed frontier models by a huge margin. Of course, I turned it down. But what genuinely scares me is knowing someone else will eventually take the deal.
3
17
12,200
A big shoutout to the @opensea team, especially @OliverMaroney, Nathan and the whole engineering team, who are building a rich @ensdomains experience on OpenSea. Really appreciate the work! 👏 Checkout: opensea.io/ens
10
10
3
76
8,802
My RTX Pros are coming soon… can’t wait!
DGX Sparks are still the #1 priority for me, especially 1-2 sparks. The extra hardware I have will accelerate everything. I feel the RTX PRO 6000s aren't optimized to their maximum potential, so it's a high priority as well.
6
7
1,353
jefflau.eth retweeted
Internet should NOT have KYC NO KYC on social media NO KYC on payments (Bitcoin) NO KYC on your data Vive la liberté
This quoted post is unavailable.
18
47
3
858
19,232
This makes no sense because if the bad guys are making profit out of abliterated models with black hat attacks, they can easily afford their own infra or buy kyc. The reason why KYC is frowned upon in crypto, is because: 1) Our KYC has been linked many times causing a safety risk. 2) It’s not permissionless. 3) It generally hurts those that can’t afford the infra and can’t use tech to defend themselves. I’m not sure if it should apply 1-1 to local AI, but definitely feels similar. Don’t create barriers for the good guys when it doesn’t stop the bad guys.
This quoted post is unavailable.
6
7
610
Own your keys. Own your name. Own your intelligence.
Own your intelligence. Nothing more important than this.
6
20
822
Yeah been weird seeing people advocate for 2*256gb mac studios. I thought it was clear if you’re going for a machine that has a ton of RAM you’re looking at running a bigger model because you’re not limited by your connectivity.
This quoted post is unavailable.
4
6
580
jefflau.eth retweeted
ENSv2 is live on Ethereum mainnet. The new protocol gives builders more flexibility to organise names and decide who can manage them. We’re glad to support the next chapter of ENS 💙 Every zk.money tag resolves via ENS, so you can get paid at bob.zk.money from any Ethereum wallet, privately.
ENSv2 is live on Ethereum mainnet. Register a name, explore the protocol, and build what comes next.
3
5
58
1,507
ENSv2 is live on mainnet! Migrate your names at app.ens.domains and you will eligible for this super cool commemorative NFT! ENS has been my life for the last decade and I'm so proud of this team for bringing the next version of ENS to make the foundations for the future of the internet. Let's go 🚀
ENSv2 is live on Ethereum mainnet. Register a name, explore the protocol, and build what comes next.
50
12
1
104
6,222
Over 9 years you've been able to register and renew your names. We're taking a small breather to make sure to migration and V2 release goes smoothly - get ready!
ENS's first outage, and it's because we're upgrading to v2. Pretty good record if you ask me!
7
1
31
1,509
Early local ai pioneers reminds me so much of early blockchain oss on steroids. It’s uncanny.
TensorFold 1.1.6 is out. Eight chats arrive together with the same new 28k-token system prompt. On a GB10 with Qwen3.8-27B, the last chat gets its first token after 31s instead of 201s. Every reply matches its run alone. Also in this release: - Opt-in conversation disk cache on Mac. - Qwen3.8-27B on Metal keeps decoding streams moving during long prompts. - Gemma 4 CUDA chat-cache fixes and GLM-5.3-Flash image-cache reuse on Metal. - Spark-X2.5/2.6-4B on CUDA and Metal. - Drama speech sampling controls and speaker tags. - TensorFold plan estimates CUDA model memory before downloading or loading weights. There are also tool-call, memory-admission and macOS download fixes, plus updates to the experimental DeepSeek-V4.1 engine. Huge thanks to the contributors: @jvr0x @bobclawblaw @krunkosaurus @lasse_ai And on GitHub: jschmied, wojo, ecohash-co, ajensenwaud, akol1, philip-pentatonic, gilby and BlivionIaG. Thanks to Boscoeuk on GitHub for the reports too. To everyone testing, sharing hardware access and supporting TensorFold, thank you. If I've missed anyone, I'm sorry. @MiaAI_lab @volatilemarkts @plotarmordev @petruspennanen tensorfold.dev
7
6
595
I was doing a workout and wanted to use my free timer that decided to start charging. I asked @bot to make a replacement for me. I had already started working out and by the time I needed the timer, grokbot had already finished the first version. 2 prompts later and it was virtually indistinguishable from the app i’ve been using for months. First time i’ve had ai generate a solution on the fly to a problem in real time. We really live in a wild time.
2
14
659
ENS is the way to name things on the internet and it's time to take it to V2
16
5
57
2,110
jefflau.eth retweeted
nick.isv2ready.eth
24
4
1
50
2,773
jefflau.eth retweeted
ens.isv2ready.eth ✅
15
23
5
168
18,003
Well I know what I'll be running all week...
Run DeepSeek V4.1 Flash 1017 with the full official weights on your 4x DGX Sparks, using TensorFold, Abliterated! 🔥 - Full support for switchless TP4 - 1M context, 16M total KV cache - 16 concurrent sessions with 1M each - 15 GB of free mem per Spark after load - OFFICIAL weights, no further quantization! - Needle passed at 100% depth - ABLIT is off by default. Gated. Performance tested on a switchless TP4. It should be faster with a switch. Decode speeds, non-greedy, thinking on: - 66 tok/s on prose, single stream - 96 tok/s on prose, 2 concurrent streams - 173 tok/s on prose, 16 concurrent streams Decode speeds, greedy, thinking off: - 85 tok/s on prose, single stream - 113 tok/s on prose, 2 concurrent streams - 210 tok/s on prose, 16 concurrent streams Decode speeds per stream in deep context (128k), non-greedy, thinking on: - 58.6 tok/s on prose, single stream - 83 tok/s on prose, 2 concurrent streams - 132 tok/s on prose, 8 concurrent streams Prefill: 4000+ tok/s up to 256k. This is the best model to run if you have 4x DGX Sparks, bar none! It works so well on the Sparks that it's unbelievable. Get it here: mia-ai.net/models/DeepSeek-V…
5
10
801