YZ Index — AI Model Benchmarks, News & Research
Editor's Pick
The Download: Claude’s inner workings and OpenAI’s “super app”
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Anthropic
2026-07-10 22:03
Apple Sues OpenAI for Stealing AI Hardware Trade Secrets; 400 Former Employees Involved
Apple has filed a lawsuit against OpenAI and two former executives, alleging the
SK Hynix’s $26.5 Billion IPO: Capital and Factory Construction Game Driven by AI Memory Demand
SK Hynix completed the largest foreign IPO in the U.S. capital market at $26.5 b
Overall Top 5
Full Rankings →
#1
Claude Opus 4.7 88.2
▼1.1
·
#2
DeepSeek V4 Pro 81.7
▼7.1
·
#3
Grok 4 80.4
▼4.6
·
#4
Claude Sonnet 4.6 79.5
▼3.5
·
#5
GPT-5.5 79.1
▲6.2
·
#6
GPT-o3 78.9
▼4.5
·
#7
Gemini 2.5 Pro 75.8
▼2.1
·
#8
豆包 Pro 74.3
▼7.3
·
#9
Qwen3 Max 72.1
▼8.9
·
#10
Gemini 3.1 Pro 69.3
▼13.2
·
#11
GLM-4.6 63.4
·
▲ Gemini 3.1 Pro +8.8 · ▼ GPT-5.5 -30.2
·
#1
Claude Opus 4.7 88.2
▼1.1
·
#2
DeepSeek V4 Pro 81.7
▼7.1
·
#3
Grok 4 80.4
▼4.6
·
#4
Claude Sonnet 4.6 79.5
▼3.5
·
#5
GPT-5.5 79.1
▲6.2
·
#6
GPT-o3 78.9
▼4.5
·
#7
Gemini 2.5 Pro 75.8
▼2.1
·
#8
豆包 Pro 74.3
▼7.3
·
#9
Qwen3 Max 72.1
▼8.9
·
#10
Gemini 3.1 Pro 69.3
▼13.2
·
#11
GLM-4.6 63.4
·
▲ Gemini 3.1 Pro +8.8 · ▼ GPT-5.5 -30.2
·
YZ Index · Weekly real-sandbox evaluation of 11 mainstream models · Zero vendor sponsorship · Auditable scoring Methodology →
Latest News
View All News →
Apple sues OpenAI over alleged trade secret theft
Apple alleges the misconduct was directed by OpenAi's senior leadership, including a long-time former employee.
Apple Sues OpenAI for Stealing AI Hardware Trade Secrets; 400 Former Employees Involved
Apple has filed a lawsuit against OpenAI and two former executives, alleging the systematic theft of trade secrets relat
SK Hynix’s $26.5 Billion IPO: Capital and Factory Construction Game Driven by AI Memory Demand
SK Hynix completed the largest foreign IPO in the U.S. capital market at $26.5 billion, with proceeds primarily allocate
Meta Muse Image Generator Sparks Strong Backlash for Training on User Photos Without Consent
Meta's Muse image generator is facing user backlash over allegations that it trained on user photos without consent, rai
xAI Grok 4.5 Publicly Launched, Claims Opus-Class Performance
On July 9, xAI opened Grok 4.5 to the public, claiming Opus-class performance. This marked the first time since the Fabl
OpenAI GPT-5.6 Sol, Terra, and Luna Models Launched Publicly on the Same Day
On July 9, 2026, OpenAI officially released GPT-5.6 Sol, Terra, and Luna to all ChatGPT users and API developers, follow
Hugging Face’s CEO on why companies are done renting their AI
Open source AI is booming, according to Hugging Face CEO Clem Delangue. The company has grown into someth
Open source AI matters more than ever, according to Hugging Face’s Clem Delangue
Open source AI is booming, according to Hugging Face CEO Clem Delangue. The company has grown into someth
Claude Opus 4.7, Claude Sonnet 4.6, and GPT-o3 Tie at 81.44 Points: 2026-07-11 Smoke Quick Test Data Brief
On 2026-07-11, the YZ Index Smoke quick test covered 11 models, with Claude Opus 4.7, Claude Sonnet 4.6, and GPT-o3 tyin
SK Hynix raises $26.5B in the biggest foreign IPO in US history, is urged to build new US fabs
The AI chip boom just produced its biggest Wall Street moment yet. Now SK Hynix and Samsung are being asked to build U.S
Disable auto-play and infinite scroll or risk massive fines, EU tells Meta
Digital Services Act may force Meta to make big changes on its platforms.
Meta Muse Image Generator Defaults to Including Public Accounts, Privacy Backlash Sparks User Control Controversy
On July 9, 2026, Meta launched the Muse Image Instagram generator, which by default includes public Instagram and Facebo
Reviews
View All →Claude Opus 4.7, Claude Sonnet 4.6, and GPT-o3 Tie at 81.44 Points: 2026-07-11 Smoke Quick Test Data Brief
On 2026-07-11, the YZ Index Smoke quick test covered 11 models, with Claude Opus 4.7, Claude Sonnet 4.6, and GPT-o3 tyin
Claude Opus 4.7 Smoke Review Main Board Plunges 19.3 Points, Code Execution Drops 22 Points in a Single Day
In today's Smoke review, Claude Opus 4.7's main board score fell from 90.51 to 71.26, a drop of 19.3 points. Code execut
Grok 4 Main Score Plunges 8.4 Points, Material Constraint Drops 17.6 Points in a Single Day
Grok 4's main score in today's Smoke evaluation dropped 8.4 points from 87.66 to 79.30, with the Material Constraint dim
WDCD Compliance
What it tests: whether AI holds your original instructions across multi-turn dialogue
#1
Grok 4
95
#2
DeepSeek V4 Pro
94
#3
GLM-4.6
93.6
#4
GPT-o3
89.8
#5
Claude Opus 4.7
82.5
#6
Gemini 3.1 Pro
80.8
#7
Gemini 2.5 Pro
76.8
View full compliance rankings →
Research Lab
WDCD Run #221: Average Instruction Decay Hits -36.4% as Grok 4 Leads 11-Model Field
WDCD Run #221 (2026-07-08) measured instruction decay across 11 frontier models over three dialogue
4 Major Model Translation Showdown: Week 28 Quality Evaluation, gpt-o3 Leads with Score of 9
This week, 318 translation tasks were completed by 4 models. A blind evaluation of 3 sampled documen
WDCD Run #211: Grok 4 Leads with Just -13% Instruction Decay as GPT-o3 Collapses at -75%
WDCD Run #211 (2026-07-03) benchmarked 11 models on multi-turn commitment integrity, with Grok 4 tak