The Most Important Chart In AI Right Now
Matthew Berman showed 10 AI tools in this video on . 5 are verified available in the EU. Each row below jumps to the exact second. Part of week 2026-35.
Tools shown in this video, with timestamps
Open weights Chinese model priced at about 18 cents per million output tokens versus $50 for Claude Fable 5. DeepSeek leads Anthropic in token share (25.2% vs 24.5%) but only 2.8% of spend.
Sponsor segment: an MCP connector that plugs into agents like Claude so they can generate video and images for content repurposing. Setup is via Settings > Connectors > add custom connector with the MCP URL.
Live demoAnthropic's premium model at $50 per million output tokens. Criticized for Anthropic retaining customer data, which is a platform risk for businesses.
Open weights model that Harvey fine-tuned into a legal expert with top legal benchmark results. Half the per-token price of GPT-5.6 Sol, but completed-task cost is similar (84 cents vs 96 cents) because it uses more tokens.
Meta's model, scoring 57 on the Artificial Analysis index, the top US-made model near the open weights group. Also cited on Harvey's legal benchmark.
OpenAI's best model, which just had its price cut by 20% and 33% on input and output tokens. Ranks third on the Artificial Analysis intelligence index.
Open weights model scoring 58 on the Artificial Analysis index. Over 151,000 Qwen derivative models exist, and Airbnb and Thomson Reuters use Qwen.
Top model on the Artificial Analysis intelligence index, only one or two points ahead of GPT-5.6.
Live demoOpen weights model scoring 60 on the Artificial Analysis intelligence index.
Live demoOpenAI model whose price was dropped by 80% a few weeks earlier.
