AI benchmark coverage: what MMLU, GPQA, SWE-bench, Terminal-Bench and ARC-AGI actually measure, who leads each one, and how to read a vendor’s claim critically.
We priced all 29 models in our database and split them open vs proprietary. The 5 cheapest are all open-weight; the 5 most expensive all proprietary. The typical gap: 39×.
We combined live API pricing with the Artificial Analysis Intelligence Index to rank AI models by intelligence per dollar. The price spread is 114× — but value tells a very different story.
Top open-source LLMs in 2026 ranked by capability + the exact hardware you need to run each locally. Llama 3 405B, Qwen 2.5 72B, DeepSeek V3, Mistral Large 2.
The complete VRAM cheat sheet for every major open LLM in 2026 — at every common quantization level — plus a matrix showing which models fit on 12, 16, 24, 32, 48, and 80 GB GPUs.
