Combining 3060 and 4070 Super for local AI: feasibility and setup
HsSekhon · reddit · 2026-08-17
A user inquires about the feasibility of combining an RTX 3060 and an RTX 4070 Super for local AI inference, questioning performance impact due to generation gaps and power requirements for a 750W PSU. The post also seeks advice on integrating this setup with OpenCode, specifically asking if LM Studio should be run as a server, and requests recommendations for generic coding models suitable for a 24GB VRAM configuration.
More from Infra
- Groq Raises $350M at $3.5B Valuation After Nvidia Deal — dinabass · 2026-08-17
- llama.cpp releases v0.1.0, adopts semantic versioning — Warrenio · 2026-08-17
- DeepSeek V4 Flash tested on Mac Studio: Impressive quality, high RAM demand — pj-frey · 2026-08-17
- Community discussion: EXL3 fades due to lack of RAM overflow support — silenceimpaired · 2026-08-17
- US AI expansion hits power bottleneck; SpaceX plans orbital compute with Starmind — XFreeze · 2026-08-17
- Use Flashinfer for VLLM on Ampere Hardware — mayo551 · 2026-08-17