Feedback
Top Charts
Today News
Basics
AIB — Compare AI Performance, Pricing, Rankings
Newsroom
Latest AI News
All news →
Latest
안전·윤리
Education
Policy
Medical
Legal
Finance
Coding
Stanford Launches Biomni Biomedical Research AI Agent
Stanford researchers released Biomni, an open-source AI agent designed to handle complex biomedical research workflows.
Today
Today
AI Identifies Hidden Warning Signs Before Major Earthquakes
Unsupervised machine learning identifies seismic warning patterns appearing weeks to months before major earthquakes.
Today
Today
Six Google Sheets Tasks Automated by Gemini
Gemini achieved a 70.48% score on the 912-question SpreadsheetBench V1 - Full benchmark in March 2026.
Today
Today
Greece Debates National AI Act Implementation Framework
Greece begins parliamentary debate on national legislation to implement the European Union's AI Act framework.
Today
Today
Anthropic Charges Free User $16.6 Million In Billing Error
Korean Claude user reports receiving $16.6 million in erroneous invoices via official channels.
Today
Today
Apple Sues OpenAI Over Alleged Theft of Trade Secrets
Apple sued OpenAI on July 10, 2026, alleging widespread theft of trade secrets and confidential hardware engineering data.
Today
Today
AI Industry Shifts Toward Efficiency and Trust
OpenAI released the GPT-5.6 model, prioritizing computational efficiency over raw performance gains
Today
Today
Interactive Tool Visualizes AI Instruction Decay Under Load
Gábor Mészáros released an interactive tool visualizing how high context load causes AI instruction decay.
Today
Today
VeriTest Restructures AI Operations to Boost Consulting Methodology
VeriTest executive Lena shuts down standalone AI evaluation platform to improve operational efficiency.
Today
Today
Claude Users Report Increased Model Preachiness and Inconsistency
Users report increased difficulty with Claude, citing overly defensive and inconsistent safety refusals for creative and benign queries.
Today
Today
Top AI Ranking
AI Model Ranking
All rankings →
Intelligence
Coding
User Preference
Image Preference
Lies
Speed
1
Anthropic
Claude Fable 5
59.9
+24.4
2
OpenAI
GPT-5.6 Sol
58.9
+23.4
3
Anthropic
Claude Opus 4.8
55.7
+20.3
4
OpenAI
GPT-5.6 Terra
55.0
+19.5
5
OpenAI
GPT-5.5
54.8
+19.3
Intelligence · Coding
User Preference
Lies · Speed
Intelligence (AA Intelligence Index)
1
Anthropic
Claude Fable 5
59.9
+24.4
2
OpenAI
GPT-5.6 Sol
58.9
+23.4
3
Anthropic
Claude Opus 4.8
55.7
+20.3
4
OpenAI
GPT-5.6 Terra
55.0
+19.5
5
OpenAI
GPT-5.5
54.8
+19.3
Coding (AA Coding Index)
1
OpenAI
GPT-5.6 Sol
77.4
+34.8
2
OpenAI
GPT-5.6 Terra
76.7
+34.1
3
Anthropic
Claude Fable 5
76.5
+33.9
4
OpenAI
GPT-5.5
74.9
+32.3
5
Anthropic
Claude Opus 4.8
74.3
+31.7
AI Trends
pos 51%
vs
neg 49%
Gemini 3.5 Flash
280 opinions analyzed
pos 53%
vs
neg 47%
Grok 4.5
310 opinions analyzed
pos 9%
vs
neg 91%
GPT-5.6 Sol
340 opinions analyzed
pos 15%
vs
neg 85%
Claude Opus 4.8
310 opinions analyzed
pos 8%
vs
neg 92%
Nano Banana 2 Lite
90 opinions analyzed
pos 56%
vs
neg 44%
GLM 5.2
514 opinions analyzed
Featured Research
Trending AI Research
All papers →
HF Daily Papers
Infinite Worlds with Versatile Interactions
Zelin Gao, Qiuyu Wang, Jiapeng Zhu et al.
2026-07-08
27
Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence
Shuailei Ma, Jiaqi Liao, Xinyang Wang et al.
2026-07-08
43
Continuous Audio Language Models
Rouard Simon, Orsini Manu, Roebel Axel et al.
2025-09-08
11
AutoDev: Automated AI-Driven Development
Michele Tufano, Anisha Agarwal, Jinu Jang et al.
2024-03-13
15
SkillOpt: Executive Strategy for Self-Evolving Agent Skills
Yifan Yang, Ziyang Gong, Weiquan Huang et al.
2026-05-22
254
arXiv
HF Daily Papers
UniClawBench: A Universal Benchmark for Proactive Agents on Real-World Tasks
Zhekai Chen, Chengqi Duan, Kaiyue Sun et al.
2026-07-09
OpenCoF: Learning to Reason Through Video Generation
Xinyan Chen, Ziyu Guo, Renrui Zhang et al.
2026-07-09
Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation
Yifan Zhou, Qihao Yang, Yan Li et al.
2026-07-09
Score Accuracy Along the Forward Diffusion Does Not Certify Numerical Stability in Diffusion Sampling
Yiwei Zhou
2026-07-09
MulTTiPop: A Multitrack Transcription Dataset for Pop Music
Nathan Pruyne, Benjamin Stoler, William Chen et al.
2026-07-09