leap/Q8_0.json

{
  "inference_type": "llama.cpp/text-to-text",
  "schema_version": "1.0.0",
  "load_time_parameters": {
    "model": "../LFM2-2.6B-Transcript-Q8_0.gguf"
  },
  "generation_time_parameters": {
    "sampling_parameters": {
      "temperature": 0.3,
      "min_p": 0.15,
      "repetition_penalty": 1.05
    }
  }
}