Three major Chinese AI labs have released powerful open-weight Mixture-of-Experts (MoE) models: Moonshot AI's Kimi K3, DeepSeek V4 Pro, and Zhipu AI's GLM-5.2. These models are designed for long-horizon coding and agent tasks, featuring million-token context windows. Kimi K3 leads in benchmark performance and is positioned as the first open 3T-class model, though its weights are not yet publicly available and serving costs are higher. DeepSeek V4 Pro offers the most cost-effective serving and is MIT-licensed, while GLM-5.2, though smaller, remains a capable option with a similar license. AI
IMPACT These models advance open-source AI capabilities, offering alternatives for long-context tasks and potentially lowering costs for developers.
RANK_REASON Comparison of multiple open-weight AI models on benchmarks, licensing, and cost.
- Claude Fable 5
- DeepSeek
- DeepSeek V4 Pro
- Gemini 3.1 Pro
- GLM-5.2
- GPT-5.5
- GPT-5.6 Sol
- Kimi K3
- Moonshot AI
- Opus 4.8
- Zhipu AI
- Mastodon
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
