mradermacher/Lumimaid-v0.2-8B-i1-GGUF · Failed to load model in LM Studio

Hiei

Jul 27

Got this error. Am I probably doing something wrong?

{
  "title": "Failed to load model",
  "cause": "llama.cpp error: 'check_tensor_dims: tensor 'blk.0.attn_q.weight' has wrong shape; expected  5120,  5120, got  5120,  4096,     1,     1'",
  "errorData": {
    "n_ctx": 2048,
    "n_batch": 512,
    "n_gpu_layers": 10
  },
  "data": {
    "memory": {
      "ram_capacity": "31.95 GB",
      "ram_unused": "20.19 GB"
    },
    "gpu": {
      "gpu_names": [
        "NVIDIA GeForce RTX 3060"
      ],
      "vram_recommended_capacity": "12.00 GB",
      "vram_unused": "10.98 GB"
    },
    "os": {
      "platform": "win32",
      "version": "10.0.22631"
    },
    "app": {
      "version": "0.2.28",
      "downloadsDir": "D:\\SoftoData\\LMStudio"
    },
    "model": {}
  }
}```

mradermacher

Owner Jul 27

Your software is probably outdated. Try with current llama.cpp or koboldcpp and it should work.

mradermacher changed discussion status to closed Jul 27