
Seattle & big-tech exile. Investing in my family, community, and country 🇺🇸. Slowly re-making my X re-entry. CPO @Subsplash
Boise, ID
Joined March 2009
- Tweets5.9K
- Following1.7K
- Followers590
- Likes9K
Chris Balt retweeted
Replying to @JackPosobiec
My wife deleted her Instagram a few weeks ago.
She is much much much MUCH happier without it.
I highly recommend this to everyone.
Chris Balt retweeted
Say what you want about The Masters...
But there is nowhere else on planet earth where you will see a photo of humans like this where not a single one of them is looking at a phone.
Truly one of the most beautiful things.
Chris Balt retweeted
Need @AnthropicAI to let me toggle between personal and enterprise Claude account and need it yesterday.
Surely I am not the only person requiring this.
Chris Balt retweeted
optimism is the best life and longevity hack
negative thinking destroys your brain and is associated with alzheimers disease
whereas optimism is associated with longevity
you must retain your childlike optimism and be a silly goose
Chris Balt retweeted
apparently software as we know it is dead
because Susan from accounting is going to vibe code a calendly alternative while on her lunch break
I think some of you need to go outside and speak to real people
Chris Balt retweeted
The NT teaches that Baptism:
1️⃣is for Christ’s disciples
2️⃣is for those who repent of sin and believe in Christ
3️⃣marks people in union with Christ
The NT teaches that the baptized church is:
4️⃣Christ’s body
5️⃣God’s new covenant people
The baptism of infants contradicts these principles.
(Pages 781–2)
Chris Balt retweeted
Debunking The Biggest Lies Told About Charlie Kirk
(0:00) - Intro
(10:58) - On Second Amendment
(12:48) - On Empathy
(15:42) - On Gay People
(17:58) - On Trans People
(24:21) - On Racism
(28:21) - On Affirmative Action
(30:58) - On Critical Race Theory
(34:56) - On Illegal Immigration
(39:12) - On Abortion
(44:22) - Charlie’s Why
(49:48) - Liberal Threats
(50:55) - Leaving The Left
(52:29) - My Thoughts
Chris Balt retweeted
free speech. but we’ll shoot you.
something is deeply broken in the west. if we don’t fix it, eventually everyone could be ushered at gunpoint to sign the proclamations of the Truth Ministers.
The AI features in the new @NotionMail are compelling, @akothari @IvanZhao @ericaanderson @NotionHQ, but a Business plan per email address feels excessive. Can one paid account cover AI for all inboxes in the app? Anyone else in this boat?
2
2
184
I’m not sure what’s more unbelievable: The feats of engineering —again—from @SpaceX or the fact that this achievement is practically nowhere in the mainstream media.
Either way, remarkable. 👏🏻🇺🇸
1
4
477
Chris Balt retweeted
New report by leading semiconductor analyst Dylan Patel shows that DeepSeek spent over $1 billion on its compute cluster. The widely reported $6M number is highly misleading, as it excludes capex and R&D, and at best describes the cost of the final training run only.
SemiAnalysis published an analysis on DeepSeek, addressing recent claims about its cost and performance. $NVDA
The report states that the widely circulated $6M training cost for DeepSeek V3 is incorrect, as it only accounts for GPU pre-training expenses and excludes R&D, infrastructure, and other critical costs. According to their findings, DeepSeek’s total server CapEx is around $1.3B, with a significant portion allocated to maintaining and operating its GPU clusters.
The report also states that DeepSeek has access to roughly 50,000 Hopper GPUs, but clarifies that this does not mean 50,000 H100s, as some have suggested. Instead, it’s a mix of H800s, H100s, and the China-specific H20s, which NVIDIA has been producing in response to U.S. export restrictions. SemiAnalysis points out that DeepSeek operates its own datacenters and has a more streamlined structure compared to larger AI labs.
On performance, the report notes that R1 matches OpenAI’s o1 in reasoning tasks but is not the clear leader across all metrics. It also highlights that while DeepSeek has gained attention for its pricing and efficiency, Google’s Gemini Flash 2.0 is similarly capable and even cheaper when accessed through API.
A key innovation cited is Multi-Head Latent Attention (MLA), which significantly reduces inference costs by cutting KV cache usage by 93.3%. The report suggests that any improvements DeepSeek makes will likely be adopted by Western AI labs almost immediately.
SemiAnalysis also mentions that costs could fall another 5x by the end of the year, and that DeepSeek’s structure allows it to move quickly compared to larger, more bureaucratic AI labs. However, it notes that scaling up in the face of tightening U.S. export controls remains a challenge.
Readers added context they thought people might want to know
It was explicitly stated in the original DeepSeek report that the ~$6M is a calculation based on the compute time of the final training run using a typical rental rate (without claiming HW was rented). It was never claimed that this would be cost of full development or such.arxiv.org/pdf/2412.19437… p5
This is the most glorious video to see on the homepage of the People's House this morning.
Happy Liberation Day!
2
162
The @theallinpod just became required listening in the tech industry as @DavidSacks leads the charge on AI & Crypto policy. Take the bull by the horns and build!
truthsocial.com/@realDonaldT…
1
106
























