Vlalika

The same problem I've suffered for 2 days. The task is to run GGUF models of the third generation of Gemma 3, since there are no problems with Gemma 2,...

There is no task to use rtx3090, everything worked for me, but I can't go beyond Gemma 2 or other 2nd generation AI when modern models run on the same...

ggml_cuda_init: GGML_CUDA_FORCE_MMQ: yes ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes llama_load_model_from_file: using device CUDA0 (NVIDIA GeForce RTX 3090)...

I noticed that version CUDA has nothing to do with LLAMA's work. Since if we look at LM studio, their software works without problems with any version, perhaps the problem...