feat: add Arcee AI provider plugin

Add a bundled Arcee AI provider plugin with ARCEEAI_API_KEY onboarding,
Trinity model catalog (mini, large-preview, large-thinking), and
OpenAI-compatible API support.

- Trinity Large Thinking: 256K context, reasoning enabled
- Trinity Large Preview: 128K context, general-purpose
- Trinity Mini 26B: 128K context, fast and cost-efficient

.github/labeler.yml

@@ -257,6 +257,10 @@
   - changed-files:
       - any-glob-to-any-file:
           - "extensions/acpx/**"
+"extensions: arcee":
+  - changed-files:
+      - any-glob-to-any-file:
+          - "extensions/arcee/**"
 "extensions: byteplus":
   - changed-files:
       - any-glob-to-any-file:

CHANGELOG.md

@@ -72,6 +72,7 @@ Docs: https://docs.openclaw.ai
 
 - Agents/video generation: add the built-in `video_generate` tool so agents can create videos through configured providers and return the generated media directly in the reply.
 - Agents/music generation: ignore unsupported optional hints such as `durationSeconds` with a warning instead of hard-failing requests on providers like Google Lyria.
+- Providers/Arcee AI: add a bundled Arcee AI provider plugin with `ARCEEAI_API_KEY` onboarding, Trinity model catalog (mini, large-preview, large-thinking), and OpenAI-compatible API support.
 - Providers/ComfyUI: add a bundled `comfy` workflow media plugin for local ComfyUI and Comfy Cloud workflows, including shared `image_generate`, `video_generate`, and workflow-backed `music_generate` support, with prompt injection, optional reference-image upload, live tests, and output download.
 - Tools/music generation: add the built-in `music_generate` tool with bundled Google (Lyria) and MiniMax providers plus workflow-backed Comfy support, including async task tracking and follow-up delivery of finished audio.
 - Providers: add bundled Qwen, Fireworks AI, and StepFun providers, plus MiniMax TTS, Ollama Web Search, and MiniMax Search integrations for chat, speech, and search workflows. (#60032, #55921, #59318, #54648)

docs/docs.json

@@ -1231,6 +1231,7 @@
                 "pages": [
                   "providers/alibaba",
                   "providers/anthropic",
+                  "providers/arcee",
                   "providers/bedrock",
                   "providers/bedrock-mantle",
                   "providers/chutes",

docs/providers/arcee.md

@@ -0,0 +1,74 @@
+---
+title: "Arcee AI"
+summary: "Arcee AI setup (auth + model selection)"
+read_when:
+  - You want to use Arcee AI with OpenClaw
+  - You need the API key env var or CLI auth choice
+---
+
+# Arcee AI
+
+[Arcee AI](https://arcee.ai) provides access to the Trinity family of mixture-of-experts models through an OpenAI-compatible API. All Trinity models are Apache 2.0 licensed.
+
+- Provider: `arcee`
+- Auth: `ARCEEAI_API_KEY`
+- API: OpenAI-compatible
+- Base URL: `https://api.arcee.ai/api/v1`
+
+## Quick start
+
+1. Get an API key from the [Arcee platform dashboard](https://app.arcee.ai).
+
+2. Set the API key (recommended: store it for the Gateway):
+
+```bash
+openclaw onboard --auth-choice arceeai-api-key
+```
+
+3. Set a default model:
+
+```json5
+{
+  agents: {
+    defaults: {
+      model: { primary: "arcee/trinity-large-thinking" },
+    },
+  },
+}
+```
+
+## Non-interactive example
+
+```bash
+openclaw onboard --non-interactive \
+  --mode local \
+  --auth-choice arceeai-api-key \
+  --arceeai-api-key "$ARCEEAI_API_KEY"
+```
+
+This will set `arcee/trinity-large-thinking` as the default model.
+
+## Environment note
+
+If the Gateway runs as a daemon (launchd/systemd), make sure `ARCEEAI_API_KEY`
+is available to that process (for example, in `~/.openclaw/.env` or via
+`env.shellEnv`).
+
+## Built-in catalog
+
+OpenClaw currently ships this bundled Arcee catalog:
+
+| Model ref                      | Name                   | Input | Context | Cost (in/out per 1M) | Notes                                     |
+| ------------------------------ | ---------------------- | ----- | ------- | -------------------- | ----------------------------------------- |
+| `arcee/trinity-large-thinking` | Trinity Large Thinking | text  | 256K    | $0.25 / $0.90        | Default model; reasoning enabled          |
+| `arcee/trinity-large-preview`  | Trinity Large Preview  | text  | 128K    | $0.25 / $1.00        | General-purpose; 400B params, 13B active  |
+| `arcee/trinity-mini`           | Trinity Mini 26B       | text  | 128K    | $0.045 / $0.15       | Fast and cost-efficient; function calling |
+
+The onboarding preset sets `arcee/trinity-large-thinking` as the default model.
+
+## Supported features
+
+- Streaming
+- Tool use / function calling
+- Structured output (JSON mode and JSON schema)
+- Extended thinking (Trinity Large Thinking)

docs/providers/index.md

@@ -29,6 +29,7 @@ Looking for chat channel docs (WhatsApp/Telegram/Discord/Slack/Mattermost (plugi
 - [Alibaba Model Studio](/providers/alibaba)
 - [Amazon Bedrock](/providers/bedrock)
 - [Anthropic (API + Claude CLI)](/providers/anthropic)
+- [Arcee AI (Trinity models)](/providers/arcee)
 - [BytePlus (International)](/concepts/model-providers#byteplus-international)
 - [Chutes](/providers/chutes)
 - [ComfyUI](/providers/comfy)

extensions/arcee/api.ts

@@ -0,0 +1,3 @@
+export { buildArceeModelDefinition, ARCEE_BASE_URL, ARCEE_MODEL_CATALOG } from "./models.js";
+export { buildArceeProvider } from "./provider-catalog.js";
+export { applyArceeConfig, ARCEE_DEFAULT_MODEL_REF } from "./onboard.js";

extensions/arcee/index.test.ts

@@ -0,0 +1,58 @@
+import { describe, expect, it } from "vitest";
+import { resolveProviderPluginChoice } from "../../src/plugins/provider-auth-choice.runtime.js";
+import { registerSingleProviderPlugin } from "../../test/helpers/plugins/plugin-registration.js";
+import arceePlugin from "./index.js";
+
+describe("arcee provider plugin", () => {
+  it("registers Arcee AI with api-key auth wizard metadata", async () => {
+    const provider = await registerSingleProviderPlugin(arceePlugin);
+    const resolved = resolveProviderPluginChoice({
+      providers: [provider],
+      choice: "arceeai-api-key",
+    });
+
+    expect(provider.id).toBe("arcee");
+    expect(provider.label).toBe("Arcee AI");
+    expect(provider.envVars).toEqual(["ARCEEAI_API_KEY"]);
+    expect(provider.auth).toHaveLength(1);
+    expect(resolved).not.toBeNull();
+    expect(resolved?.provider.id).toBe("arcee");
+    expect(resolved?.method.id).toBe("api-key");
+  });
+
+  it("builds the static Arcee AI model catalog", async () => {
+    const provider = await registerSingleProviderPlugin(arceePlugin);
+    expect(provider.catalog).toBeDefined();
+
+    const catalog = await provider.catalog!.run({
+      config: {},
+      env: {},
+      resolveProviderApiKey: () => ({ apiKey: "test-key" }),
+      resolveProviderAuth: () => ({
+        apiKey: "test-key",
+        mode: "api_key",
+        source: "env",
+      }),
+    } as never);
+
+    expect(catalog && "provider" in catalog).toBe(true);
+    if (!catalog || !("provider" in catalog)) {
+      throw new Error("expected single-provider catalog");
+    }
+
+    expect(catalog.provider.api).toBe("openai-completions");
+    expect(catalog.provider.baseUrl).toBe("https://api.arcee.ai/api/v1");
+    expect(catalog.provider.models?.map((model) => model.id)).toEqual([
+      "trinity-mini",
+      "trinity-large-preview",
+      "trinity-large-thinking",
+    ]);
+    expect(
+      catalog.provider.models?.find((model) => model.id === "trinity-large-thinking")?.reasoning,
+    ).toBe(true);
+    expect(
+      catalog.provider.models?.find((model) => model.id === "trinity-large-thinking")
+        ?.contextWindow,
+    ).toBe(262144);
+  });
+});

extensions/arcee/index.ts

@@ -0,0 +1,39 @@
+import { defineSingleProviderPluginEntry } from "openclaw/plugin-sdk/provider-entry";
+import { applyArceeConfig, ARCEE_DEFAULT_MODEL_REF } from "./onboard.js";
+import { buildArceeProvider } from "./provider-catalog.js";
+
+const PROVIDER_ID = "arcee";
+
+export default defineSingleProviderPluginEntry({
+  id: PROVIDER_ID,
+  name: "Arcee AI Provider",
+  description: "Bundled Arcee AI provider plugin",
+  provider: {
+    label: "Arcee AI",
+    docsPath: "/providers/arcee",
+    auth: [
+      {
+        methodId: "api-key",
+        label: "Arcee AI API key",
+        hint: "API key",
+        optionKey: "arceeaiApiKey",
+        flagName: "--arceeai-api-key",
+        envVar: "ARCEEAI_API_KEY",
+        promptMessage: "Enter Arcee AI API key",
+        defaultModel: ARCEE_DEFAULT_MODEL_REF,
+        applyConfig: (cfg) => applyArceeConfig(cfg),
+        wizard: {
+          choiceId: "arceeai-api-key",
+          choiceLabel: "Arcee AI API key",
+          groupId: "arcee",
+          groupLabel: "Arcee AI",
+          groupHint: "API key",
+        },
+      },
+    ],
+    catalog: {
+      buildProvider: buildArceeProvider,
+      allowExplicitBaseUrl: true,
+    },
+  },
+});

extensions/arcee/models.ts

@@ -0,0 +1,67 @@
+import type { ModelDefinitionConfig } from "openclaw/plugin-sdk/provider-model-shared";
+
+export const ARCEE_BASE_URL = "https://api.arcee.ai/api/v1";
+
+export const ARCEE_MODEL_CATALOG: ModelDefinitionConfig[] = [
+  {
+    id: "trinity-mini",
+    name: "Trinity Mini 26B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 131072,
+    maxTokens: 8192,
+    cost: {
+      input: 0.045,
+      output: 0.15,
+      cacheRead: 0.045,
+      cacheWrite: 0.045,
+    },
+  },
+  {
+    id: "trinity-large-preview",
+    name: "Trinity Large Preview",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 131072,
+    maxTokens: 8192,
+    cost: {
+      input: 0.25,
+      output: 1.0,
+      cacheRead: 0.25,
+      cacheWrite: 0.25,
+    },
+  },
+  {
+    id: "trinity-large-thinking",
+    name: "Trinity Large Thinking",
+    reasoning: true,
+    input: ["text"],
+    contextWindow: 262144,
+    maxTokens: 16384,
+    cost: {
+      input: 0.25,
+      output: 0.9,
+      cacheRead: 0.25,
+      cacheWrite: 0.25,
+    },
+    compat: {
+      supportsReasoningEffort: false,
+    },
+  },
+];
+
+export function buildArceeModelDefinition(
+  model: (typeof ARCEE_MODEL_CATALOG)[number],
+): ModelDefinitionConfig {
+  return {
+    id: model.id,
+    name: model.name,
+    api: "openai-completions",
+    reasoning: model.reasoning,
+    input: model.input,
+    cost: model.cost,
+    contextWindow: model.contextWindow,
+    maxTokens: model.maxTokens,
+    ...(model.compat ? { compat: model.compat } : {}),
+  };
+}

extensions/arcee/onboard.ts

@@ -0,0 +1,26 @@
+import {
+  createModelCatalogPresetAppliers,
+  type OpenClawConfig,
+} from "openclaw/plugin-sdk/provider-onboard";
+import { buildArceeModelDefinition, ARCEE_BASE_URL, ARCEE_MODEL_CATALOG } from "./api.js";
+
+export const ARCEE_DEFAULT_MODEL_REF = "arcee/trinity-large-thinking";
+
+const arceePresetAppliers = createModelCatalogPresetAppliers({
+  primaryModelRef: ARCEE_DEFAULT_MODEL_REF,
+  resolveParams: (_cfg: OpenClawConfig) => ({
+    providerId: "arcee",
+    api: "openai-completions",
+    baseUrl: ARCEE_BASE_URL,
+    catalogModels: ARCEE_MODEL_CATALOG.map(buildArceeModelDefinition),
+    aliases: [{ modelRef: ARCEE_DEFAULT_MODEL_REF, alias: "Arcee AI" }],
+  }),
+});
+
+export function applyArceeProviderConfig(cfg: OpenClawConfig): OpenClawConfig {
+  return arceePresetAppliers.applyProviderConfig(cfg);
+}
+
+export function applyArceeConfig(cfg: OpenClawConfig): OpenClawConfig {
+  return arceePresetAppliers.applyConfig(cfg);
+}

extensions/arcee/openclaw.plugin.json

@@ -0,0 +1,28 @@
+{
+  "id": "arcee",
+  "enabledByDefault": true,
+  "providers": ["arcee"],
+  "providerAuthEnvVars": {
+    "arcee": ["ARCEEAI_API_KEY"]
+  },
+  "providerAuthChoices": [
+    {
+      "provider": "arcee",
+      "method": "api-key",
+      "choiceId": "arceeai-api-key",
+      "choiceLabel": "Arcee AI API key",
+      "groupId": "arcee",
+      "groupLabel": "Arcee AI",
+      "groupHint": "API key",
+      "optionKey": "arceeaiApiKey",
+      "cliFlag": "--arceeai-api-key",
+      "cliOption": "--arceeai-api-key <key>",
+      "cliDescription": "Arcee AI API key"
+    }
+  ],
+  "configSchema": {
+    "type": "object",
+    "additionalProperties": false,
+    "properties": {}
+  }
+}

extensions/arcee/package.json

@@ -0,0 +1,12 @@
+{
+  "name": "@openclaw/arcee-provider",
+  "version": "2026.4.4",
+  "private": true,
+  "description": "OpenClaw Arcee provider plugin",
+  "type": "module",
+  "openclaw": {
+    "extensions": [
+      "./index.ts"
+    ]
+  }
+}

extensions/arcee/provider-catalog.ts

@@ -0,0 +1,10 @@
+import type { ModelProviderConfig } from "openclaw/plugin-sdk/provider-model-shared";
+import { buildArceeModelDefinition, ARCEE_BASE_URL, ARCEE_MODEL_CATALOG } from "./api.js";
+
+export function buildArceeProvider(): ModelProviderConfig {
+  return {
+    baseUrl: ARCEE_BASE_URL,
+    api: "openai-completions",
+    models: ARCEE_MODEL_CATALOG.map(buildArceeModelDefinition),
+  };
+}

src/agents/model-catalog.ts

@@ -44,7 +44,7 @@ const defaultImportPiSdk = () => import("./pi-model-discovery-runtime.js");
 let importPiSdk = defaultImportPiSdk;
 let modelSuppressionPromise: Promise<typeof import("./model-suppression.runtime.js")> | undefined;
 
-const NON_PI_NATIVE_MODEL_PROVIDERS = new Set(["deepseek", "kilocode", "ollama"]);
+const NON_PI_NATIVE_MODEL_PROVIDERS = new Set(["arcee", "deepseek", "kilocode", "ollama"]);
 
 function shouldLogModelCatalogTiming(): boolean {
   return process.env.OPENCLAW_DEBUG_INGRESS_TIMING === "1";

src/commands/onboard-types.ts

@@ -12,6 +12,7 @@ export type BuiltInAuthChoice =
   | "oauth"
   | "setup-token"
   | "token"
+  | "arceeai-api-key"
   | "chutes"
   | "deepseek-api-key"
   | "openai-codex"
@@ -65,6 +66,7 @@ export type AuthChoice = BuiltInAuthChoice | (string & {});
 export type BuiltInAuthChoiceGroupId =
   | "openai"
   | "anthropic"
+  | "arcee"
   | "chutes"
   | "deepseek"
   | "google"
@@ -131,6 +133,7 @@ export type OnboardOptions = OnboardDynamicProviderOptions & {
   tokenExpiresIn?: string;
   /** API key persistence mode for setup flows (default: plaintext). */
   secretInputMode?: SecretInputMode;
+  arceeaiApiKey?: string;
   cloudflareAiGatewayAccountId?: string;
   cloudflareAiGatewayGatewayId?: string;
   customBaseUrl?: string;

src/config/io.ts

@@ -73,6 +73,7 @@ export { MissingEnvVarError } from "./env-substitution.js";
 const SHELL_ENV_EXPECTED_KEYS = [
   "OPENAI_API_KEY",
   "ANTHROPIC_API_KEY",
+  "ARCEEAI_API_KEY",
   "DEEPSEEK_API_KEY",
   "ANTHROPIC_OAUTH_TOKEN",
   "GEMINI_API_KEY",