LlmClient.listModels

Java · package io.github.muthuishere:toolnexus · SPEC §8

public static List<LlmClient.ModelInfo> listModels(LlmClient.Options opts)

public record ModelInfo(String id, String name, int contextLength) {}
// name: provider display name ("" when none); contextLength: tokens, 0 = unknown

Lists the models a provider serves, sorted by id. openai style: GET {baseUrl}/models (OpenRouter’s context_length is read when present). anthropic style: GET {baseUrl}/v1/models, following has_more / last_id. One GET per page, no retries, opts.httpClient honoured, and the same key rules as the client (API key by host). A non-2xx throws list models: HTTP <status>: ….

When to use it

A model picker or a /model command in a host: list what the configured provider offers, show the display name and context window, then build the client with the chosen model.

Why this and not the alternative

Examples

1. The smallest useful call

import com.sun.net.httpserver.HttpServer;
import io.github.muthuishere.toolnexus.*;
import java.io.OutputStream;
import java.net.InetSocketAddress;
import java.nio.charset.StandardCharsets;
import java.util.List;

public class Example {
    public static void main(String[] args) throws Exception {
        HttpServer server = HttpServer.create(new InetSocketAddress("127.0.0.1", 0), 0);
        server.createContext("/models", ex -> {
            byte[] b = "{\"data\":[{\"id\":\"zeta\"},{\"id\":\"alpha\"}]}".getBytes(StandardCharsets.UTF_8);
            ex.sendResponseHeaders(200, b.length);
            try (OutputStream os = ex.getResponseBody()) { os.write(b); }
        });
        server.start();
        try {
            List<LlmClient.ModelInfo> models = LlmClient.listModels(new LlmClient.Options()
                    .baseUrl("http://127.0.0.1:" + server.getAddress().getPort())
                    .apiKey("YOUR_KEY_HERE"));
            if (!models.get(0).id().equals("alpha")) throw new AssertionError(models);
            System.out.println("ok: " + models.stream().map(LlmClient.ModelInfo::id).toList());
        } finally {
            server.stop(0);
        }
    }
}

2. The realistic case — OpenRouter-shaped names and context lengths

import com.sun.net.httpserver.HttpServer;
import io.github.muthuishere.toolnexus.*;
import java.io.OutputStream;
import java.net.InetSocketAddress;
import java.nio.charset.StandardCharsets;

public class Example {
    public static void main(String[] args) throws Exception {
        HttpServer server = HttpServer.create(new InetSocketAddress("127.0.0.1", 0), 0);
        server.createContext("/api/v1/models", ex -> {
            byte[] b = "{\"data\":[{\"id\":\"vendor/big\",\"name\":\"Big Model\",\"context_length\":200000}]}"
                    .getBytes(StandardCharsets.UTF_8);
            ex.sendResponseHeaders(200, b.length);
            try (OutputStream os = ex.getResponseBody()) { os.write(b); }
        });
        server.start();
        try {
            for (LlmClient.ModelInfo m : LlmClient.listModels(new LlmClient.Options()
                    .baseUrl("http://127.0.0.1:" + server.getAddress().getPort() + "/api/v1")
                    .apiKey("YOUR_KEY_HERE"))) {
                if (m.contextLength() != 200000) throw new AssertionError(m);
                System.out.println("ok: " + m.name() + " (" + m.id() + "), " + m.contextLength() + " tokens");
            }
        } finally {
            server.stop(0);
        }
    }
}

3. The full surface — anthropic paging

import com.sun.net.httpserver.HttpServer;
import io.github.muthuishere.toolnexus.*;
import java.io.OutputStream;
import java.net.InetSocketAddress;
import java.nio.charset.StandardCharsets;
import java.util.List;

public class Example {
    public static void main(String[] args) throws Exception {
        HttpServer server = HttpServer.create(new InetSocketAddress("127.0.0.1", 0), 0);
        server.createContext("/v1/models", ex -> {
            boolean second = String.valueOf(ex.getRequestURI().getQuery()).contains("after_id=m-b");
            String json = second
                    ? "{\"data\":[{\"id\":\"m-c\",\"display_name\":\"C\"}],\"has_more\":false}"
                    : "{\"data\":[{\"id\":\"m-b\",\"display_name\":\"B\"},{\"id\":\"m-a\",\"display_name\":\"A\"}],"
                      + "\"has_more\":true,\"last_id\":\"m-b\"}";
            byte[] b = json.getBytes(StandardCharsets.UTF_8);
            ex.sendResponseHeaders(200, b.length);
            try (OutputStream os = ex.getResponseBody()) { os.write(b); }
        });
        server.start();
        try {
            List<LlmClient.ModelInfo> models = LlmClient.listModels(new LlmClient.Options()
                    .baseUrl("http://127.0.0.1:" + server.getAddress().getPort())
                    .style("anthropic").apiKey("YOUR_KEY_HERE"));
            List<String> ids = models.stream().map(LlmClient.ModelInfo::id).toList();
            if (!ids.equals(List.of("m-a", "m-b", "m-c"))) throw new AssertionError(ids);
            System.out.println("ok: two pages -> " + ids);
        } finally {
            server.stop(0);
        }
    }
}