Blog
Analysis, benchmarks, and insights on Chinese AI models.
Best Chinese AI Model for Your Use Case: A Decision Framework
Coding? Use DeepSeek V3.2. Long documents? Qwen3.5-Plus. Zero budget? GLM-4.7-Flash. Here's how to pick the right Chinese LLM for every scenario.
DeepSeek vs GPT for Coding: Benchmarks, Real Costs, and When Each Wins
DeepSeek V3.2 costs $0.28/M tokens vs GPT-5.2's $1.75. But GPT leads SWE-bench by 7 points. Here's when DeepSeek is worth it — and when it's not.
What Chinese LLMs Censor (And What They Don't)
85% of politically sensitive prompts get blocked by DeepSeek R1. But coding, math, and business content? Completely unaffected. Here's the full picture with data.
The Mystery 1T Model Everyone Thought Was DeepSeek V4 Turned Out to Be Xiaomi
An anonymous trillion-parameter model topped OpenRouter for 8 days. The entire AI community assumed it was DeepSeek V4. It was actually Xiaomi's MiMo-V2-Pro — and that matters more than the identity reveal.
Chinese LLM Pricing Deep Dive: The Real Cost of Running AI in 2026
Beyond the price table. What Chinese models actually cost for real workloads — with caching, tiered pricing, and the hidden gotchas.