Posted inLocal LLMs
Qwen 35B running on 12gb of VRAM in LM Studio at 120+ tokens/second. Works with Cline for 100% agentic coding.
Harnessing Qwen 35B on a Consumer GPU for Advanced AI-Driven Development In recent advancements within the AI community, enthusiasts have successfully leveraged large language models (LLMs) on consumer-grade hardware, pushing…









