Research & Papers

Technical guides, research papers, and case studies

15 Articles

Why Data Quality, Not Framework Speed, Decides Small-Model Quality

On saturated consumer hardware, you do not buy small-model quality with a faster framework. You buy it with better data and a better training objective, with numbers to back it.

Jul 15, 2026

5 min read

small modelsdata efficiencybyte-level models+6

Decode Without Changing Output Bytes

Speculative decoding and multi-token prediction cut byte-level generation latency without touching output quality. Here is what we measured on Veritate's models

Jul 14, 2026

5 min read

speculative decodingmulti-token predictionbyte-level models+5

Why Low-Bit QAT Need to Happen In Training

Below roughly four bits per weight, post-hoc QAT on byte-level models is catastrophic, and quantization-aware training recovers nearly all of the lost quality

Jul 13, 2026

5 min read

quantization-aware trainingpost-training quantizationlow-bit LLMs+5

Thinking Time, Not Parameter Count

For a model that must fit on a small device, the parameter budget is fixed. The place to spend is inference-time compute.

Jul 12, 2026

5 min read

test-time computeinference-time computesmall language models+4

Upcycling the Data Center: Doing More AI With Older Hardware

We throw away a staggering amount of perfectly useful compute. Here is why a lot of AI and cloud work can run on older, upcycled hardware, and where it honestly cannot.

Jul 11, 2026

7 min read

data center e-wastesustainable computinghardware lifecycle+4

Byte-Level Language Modeling Without a Tokenizer

Why our Veritate models read raw bytes instead of tokens, what the 256-symbol vocabulary costs, and what our 85M and 800M models tell us about the trade.

Jul 1, 2026

5 min read

byte-level language modeltokenizer-freelanguage model research+3

Open Weights and the Future of Private AI

Open-weight models let you run capable AI on hardware you control. Here is why I think that matters more than any leaderboard, for privacy, cost, and independence.

Jun 29, 2026

7 min read

open weightsopen source aiprivate ai+6

Against Automating Everything: Where AI Belongs

We're not against AI. We are against the reflex to point it at everything. Here is where automation earns its place...

Jun 29, 2026

8 min read

ai automationai skepticismAutomation+4

Activation Sparsity for Efficient Inference

You cannot retrofit activation sparsity onto a finished model, but a short, cheap retrain installs it, and the savings compose with weight pruning.

Jun 11, 2026

5 min read

activation sparsityenergy-efficient inferenceReLUfication+4

Why the Internet Forgot How to Be Efficient

We ship megabytes of script to display a paragraph of text, and we call it progress. I think we forgot that efficiency was ever a virtue.

Jun 11, 2026

8 min read

web bloatefficiencysustainable computing+3

The Cost of "Free" AI: Who Pays for Inference

Free AI is not a gift. Someone pays for every query, in energy, in data, in dependency. Here is who, and why I'd rather pay honestly for compute I control

Jun 3, 2026

8 min read

cost of free aiai inferenceai energy use+3

React2Shell Explained: Carpathian's Incident Response

CVE-2025-55182 React2Shell: Critical Zero-Day RCE. Real Cyber Attack Analysis and Threat Detection, IOCs, Remediation Steps. React & Next.js Security.

Dec 17, 2025

10 min read

Critical RCE in React Server Components

React2Shell (CVE-2025-55182): Critical RCE in React Server Components. Patch to React 19.0.1+ and Next.js 15.0.5+ immediately.

Dec 15, 2025

9 min read

AI Pen-Testing: What You Need to Know

Anthropic reports the first documented AI cyberattack. Chinese state-sponsored hackers use Claude AI to perform reconnaissance, exploitation, and data theft

Nov 21, 2025

24 min read

Rethinking AI Thinking

Can artificial intelligence truly think, or is it simply an extraordinarily sophisticated pattern-matching system? The people who succeed in the AI era won't be those who blindly adopt every new AI tool or those who resist all AI assistance. They'll be the ones who understand exactly what AI can and cannot do.

Nov 20, 2025

13 min read