DeepSeek-OCR is a vision-language model that can perform token-efficient OCR.
DeepSeek-V4-Pro is a frontier Mixture-of-Experts model with a large context window and three reasoning modes.
Based on DeepSeek R1 because OpenCode tries to verify on the registry for tool compatibility
Quantized version of DeepSeek-R1-32B optimized for tool usage with Cline / Roo Code and complex problem solving.
DeepSeek-V2.5-1210 is an upgraded version of DeepSeek-V2.5, offering enhanced mathematical, coding, writing, and reasoning capabilities.