9 stable releases

8.0.0 Dec 31, 2021
7.11.0 Dec 3, 2021
7.10.0 Nov 1, 2021
7.9.0 Sep 29, 2021
0.0.2 Nov 19, 2020

#460 in Text processing

24 downloads per month


95K SLoC

Clarifai logo

Clarifai Rust gRPC Client

This is the official Clarifai gRPC Rust client for interacting with our powerful recognition API. Clarifai provides a platform for data scientists, developers, researchers and enterprises to master the entire artificial intelligence lifecycle. Gather valuable business insights from images, video and text using computer vision and natural language processing.

crates.io Run tests


Add these dependencies to Cargo.toml: clarifai_grpc, protobuf and grpcio.

clarifai_grpc = "*"
grpcio = "0.6.0"
protobuf = "2.0"


This library doesn't use semantic versioning. The first two version numbers (X.Y out of X.Y.Z) follow the API (backend) versioning, and whenever the API gets updated, this library follows it.

The third version number (Z out of X.Y.Z) is used by this library for any independent releases of library-specific improvements and bug fixes.

Getting started

Construct the V2Client object using which you'll access all the Clarifai API functionality, and a CallOption object that will be used for authentication.

use grpcio::{CallOption, MetadataBuilder};
use protobuf::{RepeatedField, SingularPtrField};

use clarifai_grpc::clarifai_channel;
use clarifai_grpc::grpc::resources;
use clarifai_grpc::grpc::service;
use clarifai_grpc::grpc::service_grpc::V2Client;
use clarifai_grpc::grpc::status_code::StatusCode;

let client = V2Client::new(clarifai_channel::grpc());

// Setup authentication.
let auth = "Key YOUR_CLARIFAI_API_KEY_OR_PAT".to_string();

let mut builder = MetadataBuilder::with_capacity(1);
builder.add_str("Authorization", &auth).unwrap();
let metadata = builder.build();
let call_opt = CallOption::default().headers(metadata);

On Windows and macOS gRPC requires explicitly setting the root of trust for SSL. One way to do this is by setting the GRPC_DEFAULT_SSL_ROOTS_FILE_PATH environmental variable. To do this on macOS use:

curl -Lo roots.pem https://raw.githubusercontent.com/grpc/grpc/master/etc/roots.pem

On Windows use:

@powershell -NoProfile -ExecutionPolicy unrestricted -Command ^
    (new-object System.Net.WebClient).Downloadfile( ^
        'https://raw.githubusercontent.com/grpc/grpc/master/etc/roots.pem', ^

See more here.

Predict concepts in an image:

// This is a publicly available model.
const GENERAL_MODEL_ID: &str = "aaa03c23b3724a16a56b629203edc62c";

let request = service::PostModelOutputsRequest {
    model_id: GENERAL_MODEL_ID.to_string(),
    inputs: RepeatedField::from(vec![resources::Input {
        data: SingularPtrField::some(resources::Data {
            image: SingularPtrField::some(resources::Image {
                url: "https://samples.clarifai.com/dog2.jpeg".to_string(),
let response = client
    .post_model_outputs_opt(&request, call_opt)

let status = response.get_status();
if status.get_code() != StatusCode::SUCCESS {
    println!("Failure response:");
    println!("\t{:?}", status.get_code());
    println!("\t{}", status.get_description());
    println!("\t{}", status.get_details());

println!("Predicted concepts:");
for concept in response.get_outputs()[0].get_data().get_concepts() {
    println!("\t{}: {}", concept.get_name(), concept.get_value());


~652K SLoC