SeededModel
This is Precursor to a Model. bytes: here is the size of the model in a file local on your computer. ram: is the amount of ram usage when the model is loaded into memory.
Properties​
Name | Type | Description | Notes |
---|---|---|---|
var_schema | EmbeddedModelSchema | [optional] | |
version | str | this is a version of the model. | |
created | GroupedTimestamp | ||
name | str | This is an Optional Name of the Model. | |
description | str | An Optional Description of the model itself. | [optional] |
cloud | bool | This will inform the user if this was a model that is hosted in the cloud | |
type | ModelTypeEnum | ||
usage | ModelUsageEnum | ||
bytes | ByteDescriptor | [optional] | |
ram | ByteDescriptor | [optional] | |
quantization | str | quantization is a string like: q8f16_0, q4f16_1, etc... | [optional] |
foundation | ModelFoundationEnum | [optional] | |
downloaded | bool | This is an optional bool to let us know if this model has been downloaded locally. | [optional] |
unique | str | This is the unique model name used to load the model. | [optional] |
parameters | float | This is the number of parameters in terms of billions. | [optional] |
provider | ExternalMLProviderEnum | [optional] | |
cpu | bool | This is an optional bool that is optimized for CPU usage. | [optional] |
max_tokens | ModelMaxTokens | [optional] | |
custom | bool | This is reserved to custommly registed models. | [optional] |