@basetenlabs/client - v0.2.0
    Preparing search index...

    Type Alias ModelTRTLLMQuantizationType

    ModelTRTLLMQuantizationType:
        | "no_quant"
        | "weights_int8"
        | "weights_kv_int8"
        | "weights_int4"
        | "weights_int4_kv_int8"
        | "smooth_quant"
        | "fp8"
        | "fp8_kv"
        | "fp8_mlp_only"
        | "fp4"
        | "fp4_kv"
        | "fp4_mlp_only"