LM Studio 0.3.7: DeepSeek R1 and KV Cache Quantization
AI Impact Summary
LM Studio has released version 0.3.7, introducing support for DeepSeek R1 distilled models and KV cache quantization for llama.cpp models. This allows users to leverage the capabilities of DeepSeek R1, a SOTA open source reasoning model, while also improving the efficiency of running llama.cpp models through KV cache quantization. This update enhances the LM Studio platform with a new model and optimization technique.
Affected Systems
- Date
- Date not specified
- Change type
- capability
- Severity
- medium