Open peter-ch opened 4 months ago
After I installed CUDA 12.5 (upgraded from 12.1) LM Studio stopped working with the GPU. It says:
"llama.cpp error: 'error loading model architecture: unknown model architecture: 'gemma2''"
{ "memory": { "ram_capacity": "79.92 GB", "ram_unused": "69.27 GB" }, "gpu": { "gpu_names": [ "NVIDIA GeForce RTX 2080 Ti", "NVIDIA GeForce GTX 1080" ], "vram_recommended_capacity": "19.00 GB", "vram_unused": "16.89 GB" }, "os": { "platform": "win32", "version": "10.0.19045" }, "app": { "version": "0.2.27", "downloadsDir": "C:\LLMs" }, "model": {} }
After I installed CUDA 12.5 (upgraded from 12.1) LM Studio stopped working with the GPU. It says:
"llama.cpp error: 'error loading model architecture: unknown model architecture: 'gemma2''"
{ "memory": { "ram_capacity": "79.92 GB", "ram_unused": "69.27 GB" }, "gpu": { "gpu_names": [ "NVIDIA GeForce RTX 2080 Ti", "NVIDIA GeForce GTX 1080" ], "vram_recommended_capacity": "19.00 GB", "vram_unused": "16.89 GB" }, "os": { "platform": "win32", "version": "10.0.19045" }, "app": { "version": "0.2.27", "downloadsDir": "C:\LLMs" }, "model": {} }