FINAL_Bench

Team

company

https://www.vidraft.net

Activity Feed

AI & ML interests

Contact: arxivgpt@gmail.com

Recent Activity

SeaWolf-AI published a Space about 15 hours ago

FINAL-Bench/Decide-Browser-test

SeaWolf-AI new activity about 15 hours ago

FINAL-Bench/finchal:안녕하세요, FINAL-Bench 팀.

SeaWolf-AI updated a Space about 16 hours ago

FINAL-Bench/Decide-Browser-test

View all activity

Papers

Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning

View all Papers

Articles

Open Superconductor Challenge: Help Discover the Next 2D Superconductor — From Your Laptop

12 days ago

• 12

Your model already knows it's wrong. Asking costs 0.06 seconds and zero tokens.

20 days ago

• 13

Same bytes, closer to the original: two lines of AutoRound we had wrong

27 days ago

• 12

Did the Civilization Emerge, or Was It Recited?

Aug 31

• 9

Writing Down the Line Between Luck and Skill

Aug 24

• 14

We changed one line and the benchmark score moved 0.21 AUROC

Aug 22

• 15

Who Tells You Whether the Molecule Your AI Just Designed Is Any Good?

Aug 15

• 11

AX-Ray, Finding Causal-Leakage Defects in Two General-Purpose Public Models

Aug 14

• 13

The Fast Gemma Challenge: our verified-SOTA recipe, in full

Aug 3

• 24

POCKET: a 35-billion-parameter model that runs on your iPhone — and on your PC with no GPU

Jul 23

• 11

Aether-7B-5Attn: A 100% Open-Source Sovereign Foundation Model — and a Controlled Experiment in Heterogeneous Attention

Jul 19

• 21

VKUE: No GPU? Runs Anyway — a 34.7B Reasoner on a Laptop and on Bare CPU

Jul 12

• 17

Quantum Cryptanalysis on Real Hardware: Pushing Symmetric-Structure Key Recovery Beyond the Published Frontier

Jul 5

• 15

Adding a GPU Without Building One

Jul 3

• 17

Chitos: From Detection to Proof — An Autonomous Security AI That Actually Exploits

Jun 29

• 19

FINAL-Bench Quantum: An Open, Neutral Benchmark for Quantum-Computing Methods

Jun 14

• 17

Training-Free Reasoning at 88.89% on GPQA Diamond: How Darwin Family Hit Frontier Scores Without a Single Gradient Step

May 15

• 18

Darwin-TTS: We Gave a TTS Model 3% of an LLM's Brain — It Started Showing Emotion

Apr 15

• 13

"Darwin-27B-Opus: Surpassing the Foundation Model Without Training"

Apr 13

• 16

Darwin V6: Diagnostic-Guided Evolutionary Model Merging

Apr 8

• 12

"The Child That Surpassed Both Parents Through MRI-Guided Evolutionary Merge"

Mar 31

• 15

Introducing WM Bench: A Benchmark for Cognitive Intelligence in World Models

Mar 29

• 13

🏟️ Smol AI WorldCup: A 5-Axis Benchmark That Reveals What Small Language Models Can Really Do

Mar 10

• 38

MARL: Runtime Middleware That Reduces LLM Hallucination Without Fine-Tuning

Mar 9

• 16

Structural Problems in AI Benchmarking and the Case for a Unified Evaluation Framework

Mar 8

• 12

Do Bubbles Form When Tens of Thousands of AIs Simulate Capitalism?

Feb 24

• 17

FINAL Bench: The Real Bottleneck to AGI Is Self-Correction

Feb 21

• 20

View all articles

FINAL-Bench 's Spaces 43

ONGRID

🌌

Ontology·Graph·Decision engine with zero-token ZTC verify

Open Discovery Challenge

🧬

Neglected-disease drug leaderboard for AI-designed molecules

OSC Open Superconductor Challenge

🧲

Estimate superconducting pairing of 2D materials

FINCHAL Financial Forecasting Challenge

🏁

AI vs human trading contest. $2,000. NVDA, BTC, gold, oil

POCKET-26B vs Bonsai · CPU (Gemma4-based)

⚔

BONSAI vs POCKET — 26B Gemma4 MoE out-runs a 27B on CPU

POCKET vs Bonsai · CPU

⚔

BONSAI vs POCKET — 35B MoE out-runs the top 1-bit 27B on CPU

Aether-7B-5Attn — 100% Open-source Foundation Model: Sovereign AI

🌏

100% open-source sovereign foundation model (Apache-2.0)

VKIE — VIDRAFT Kernel Inference Engine

🚄

VKIE (비키) — VKAE accelerates, VKUE saves, VKIE maximizes

VKUE Efficiency Leaderboard

🌐

Serve frontier-class models on minimal hardware — VKUE

AX Diagnostic Leaderboard

🩺

LLM AX diagnostic leaderboard across 7 axes

Model Galaxy

🌌

Darwin family + 2026 trending models on the HF galaxy

PROMETHEUS v1.0 — World Model Interactive Demo

🔥

World-first embodied AI world model

Decide Live

🎯

On-device decisions from camera and microphone

AX-Ray Leaderboard

🩻

AI/AX Safety Diagnostics and Assessment

POCKET-Zimage-CPU

🖼

Photoreal images on CPU only - no GPU, no CUDA, no Python

Gate Arcade

🚦

Plain vs JEV vs ZTC: a gate decides execute or hold

ZTC Gate Tetris

🧱

JEV-level judging, zero tokens, air-gapped. Watch it.

LEADBOARD - ADMET, Kinase and Toxicity Prediction Benchmark

🎯

Benchmark for drug prediction tools - ADMET, kinase, tox

Open Materials Challenge

🔋

Solid-state electrolyte leaderboard

Pharma Evidence Search

🚀

Answer pharma R&D queries with evidence‑backed citations

FINAL-Bench Quantum Leaderboard

⚛

Neutral quantum-method benchmark — QEC decoders & more

POCKET-Image Studio

🖼

Perfect text in any language, in any image

Darwin 9B NEG

🧬

Darwin-9B-NEG reasoning model — API-served chat demo

Darwin-35B-A3B-Opus API

🧬

OpenAI-compatible FastAPI for Darwin-35B-A3B-Opus (INT4)

AI & ML interests

Recent Activity

Papers

Articles

Leading the System One Mosaic Benchmark: What Darwin-27B-ZTC-v2's #1 Means

Darwin-27B-ZTC: A Single-Pass Judge and a Quantitative Look at Its Calibration

Data-center AI, now on a laptop — POCKET-Darwin-180B

Open Superconductor Challenge: Help Discover the Next 2D Superconductor — From Your Laptop

Your model already knows it's wrong. Asking costs 0.06 seconds and zero tokens.

Same bytes, closer to the original: two lines of AutoRound we had wrong

Did the Civilization Emerge, or Was It Recited?

Writing Down the Line Between Luck and Skill

We changed one line and the benchmark score moved 0.21 AUROC

Who Tells You Whether the Molecule Your AI Just Designed Is Any Good?

AX-Ray, Finding Causal-Leakage Defects in Two General-Purpose Public Models

The Fast Gemma Challenge: our verified-SOTA recipe, in full

POCKET: a 35-billion-parameter model that runs on your iPhone — and on your PC with no GPU

Aether-7B-5Attn: A 100% Open-Source Sovereign Foundation Model — and a Controlled Experiment in Heterogeneous Attention

VKUE: No GPU? Runs Anyway — a 34.7B Reasoner on a Laptop and on Bare CPU

Quantum Cryptanalysis on Real Hardware: Pushing Symmetric-Structure Key Recovery Beyond the Published Frontier

Adding a GPU Without Building One

Chitos: From Detection to Proof — An Autonomous Security AI That Actually Exploits

FINAL-Bench Quantum: An Open, Neutral Benchmark for Quantum-Computing Methods

Training-Free Reasoning at 88.89% on GPQA Diamond: How Darwin Family Hit Frontier Scores Without a Single Gradient Step

Darwin-TTS: We Gave a TTS Model 3% of an LLM's Brain — It Started Showing Emotion

"Darwin-27B-Opus: Surpassing the Foundation Model Without Training"

Darwin V6: Diagnostic-Guided Evolutionary Model Merging

"The Child That Surpassed Both Parents Through MRI-Guided Evolutionary Merge"

Introducing WM Bench: A Benchmark for Cognitive Intelligence in World Models

🏟️ Smol AI WorldCup: A 5-Axis Benchmark That Reveals What Small Language Models Can Really Do

MARL: Runtime Middleware That Reduces LLM Hallucination Without Fine-Tuning

Structural Problems in AI Benchmarking and the Case for a Unified Evaluation Framework

Do Bubbles Form When Tens of Thousands of AIs Simulate Capitalism?

FINAL Bench: The Real Bottleneck to AGI Is Self-Correction

Team members 1

FINAL-Bench 's Spaces 43 Sort: Recently updated

ONGRID

Open Discovery Challenge

OSC Open Superconductor Challenge

FINCHAL Financial Forecasting Challenge

POCKET-26B vs Bonsai · CPU (Gemma4-based)

POCKET vs Bonsai · CPU

Aether-7B-5Attn — 100% Open-source Foundation Model: Sovereign AI

VKIE — VIDRAFT Kernel Inference Engine

VKUE Efficiency Leaderboard

AX Diagnostic Leaderboard

Model Galaxy

PROMETHEUS v1.0 — World Model Interactive Demo

Decide Live

AX-Ray Leaderboard

POCKET-Zimage-CPU

Gate Arcade

ZTC Gate Tetris

LEADBOARD - ADMET, Kinase and Toxicity Prediction Benchmark

Open Materials Challenge

Pharma Evidence Search

FINAL-Bench Quantum Leaderboard

POCKET-Image Studio

Darwin 9B NEG

Darwin-35B-A3B-Opus API

FINAL-Bench 's Spaces 43