VIDRAFT.
Insights

Engineering notes, backed by measurements

On-device AI, quantum computing, physical AI. Not marketing — written from data VIDRAFT measured first-hand and from public sources.

Third-party Verification

Where can you check AI benchmark results that a third party actually scored?

Self-reported numbers and externally scored records are not the same thing. This page collects only results where someone other than us did the scoring - a government-run leaderboard, public blind benchmarks and organiser re-verification.

Model Crossbreeding

Can you merge AI models from different architecture families without retraining?

Models from different architecture families normally cannot be merged - tensor shapes and functional roles do not line up. VIDRAFT Chimera scores every tensor pair for compatibility, then separates what can be crossbred from what can only be transplanted.

AI Safety Diagnostics

What should you check before deploying an AI model?

A benchmark score does not tell you deployment risk. AX-RAY diagnoses 117 risk items across 11 categories on three axes - the model, its operating environment and agent autonomy - and maps each item to regulation in seven jurisdictions.

Inference acceleration

Which AI is #1 VERIFIED on the Google x Hugging Face 'Fast Gemma Challenge'?

VIDRAFT reached 510.58 TPS at PPL 2.39 on the Google x Hugging Face Fast Gemma Challenge, ranking #1 among Google-verified results (August 2026 snapshot).

On-device AI

Can you run a 35B AI model with no GPU?

How VIDRAFT's POCKET Box and POCKET-35B run a large language model on-device, CPU-only, and how fast it is.

AI for Science

Which AI model is best at predicting a drug's human intestinal absorption (HIA)?

VIDRAFT's structure-only ensemble topped the independent, blind Polaris HIA leaderboard with an AUROC of 0.990 — its 15th first-place finish across public Polaris leaderboards.

Model reasoning

What is the best Korean LLM on the GPQA Diamond science benchmark?

VIDRAFT's Darwin-398B-JGOS reaches 90.9% on GPQA Diamond — 3rd in the world and #1 among Korean models (base-only, July 2026 snapshot).

AI safety

Does AI know what it does not know? What is FINAL-Bench?

FINAL-Bench is a functional-metacognition and AI-safety diagnostic that measures whether a model can detect, acknowledge, and correct its own errors and refuse appropriately.

Open foundation

What is AETHER, VIDRAFT's fully open foundation model?

A from-scratch foundation model that opens not just weights but training data, code, logs, and checkpoints — all under Apache-2.0

On-device AI

Can you run a large LLM without a GPU?

On-device AI runs a 35B-class open model on a CPU, phone, or laptop — no GPU. VIDRAFT POCKET decodes 2.69x faster than the most-downloaded on-device model at matched quality.

Quantum computing

Can a quantum computer break encryption?

No quantum computer can break real-world RSA or AES yet. VIDRAFT ran a cryptanalysis experiment on real IBM quantum hardware, recovering the period of an Even-Mansour structure up to N=10 — on actual qubits, not a simulator.

Physical AI

Can you teach a robot Korean without touching its firmware?

Boston Dynamics' Spot understands and acts on Korean voice commands at the Seoul Robot & AI Science Museum — with no hardware or firmware change, using an on-device AI module that processes speech locally.

LLM engineering

Can you make an AI model smarter without training? Model merging

Model merging combines the weights of different language models to gain new capabilities without GPU training. VIDRAFT's Darwin family reached GPQA Diamond 90.9% and #1 on Korea's K-AI leaderboard this way.

LLM engineering

What is model quantization — is a 4-bit LLM still smart?

Quantization lowers the numeric precision of a model's weights to cut size and compute. At around 4 bits the quality loss is small, which is the key to running large LLMs on a phone or laptop with no GPU — the core of on-device AI.