diff --git a/leap/Q4_0.json b/leap/Q4_0.json new file mode 100644 index 0000000..20185c8 --- /dev/null +++ b/leap/Q4_0.json @@ -0,0 +1,14 @@ +{ + "inference_type": "llama.cpp/text-to-text", + "schema_version": "1.0.0", + "load_time_parameters": { + "model": "../LFM2-350M-Extract-Q4_0.gguf" + }, + "generation_time_parameters": { + "sampling_parameters": { + "temperature": 0.3, + "min_p": 0.15, + "repetition_penalty": 1.05 + } + } +} \ No newline at end of file diff --git a/leap/Q4_K_M.json b/leap/Q4_K_M.json new file mode 100644 index 0000000..3af80d2 --- /dev/null +++ b/leap/Q4_K_M.json @@ -0,0 +1,14 @@ +{ + "inference_type": "llama.cpp/text-to-text", + "schema_version": "1.0.0", + "load_time_parameters": { + "model": "../LFM2-350M-Extract-Q4_K_M.gguf" + }, + "generation_time_parameters": { + "sampling_parameters": { + "temperature": 0.3, + "min_p": 0.15, + "repetition_penalty": 1.05 + } + } +} \ No newline at end of file diff --git a/leap/Q5_K_M.json b/leap/Q5_K_M.json new file mode 100644 index 0000000..5e1ece8 --- /dev/null +++ b/leap/Q5_K_M.json @@ -0,0 +1,14 @@ +{ + "inference_type": "llama.cpp/text-to-text", + "schema_version": "1.0.0", + "load_time_parameters": { + "model": "../LFM2-350M-Extract-Q5_K_M.gguf" + }, + "generation_time_parameters": { + "sampling_parameters": { + "temperature": 0.3, + "min_p": 0.15, + "repetition_penalty": 1.05 + } + } +} \ No newline at end of file diff --git a/leap/Q8_0.json b/leap/Q8_0.json new file mode 100644 index 0000000..92b71d1 --- /dev/null +++ b/leap/Q8_0.json @@ -0,0 +1,14 @@ +{ + "inference_type": "llama.cpp/text-to-text", + "schema_version": "1.0.0", + "load_time_parameters": { + "model": "../LFM2-350M-Extract-Q8_0.gguf" + }, + "generation_time_parameters": { + "sampling_parameters": { + "temperature": 0.3, + "min_p": 0.15, + "repetition_penalty": 1.05 + } + } +} \ No newline at end of file