feat: add Morpheus Inference API provider integration

Morpheus is a decentralized inference marketplace that provides FREE access to AI models during Open Beta. The API is fully OpenAI-compatible. This integration adds: - Complete model catalog with 14 models: - Flagship: Qwen3 Coder 480B, Hermes 3 Llama 405B, GPT OSS 120B - Reasoning: Kimi K2 Thinking, GLM 4.7 Thinking, Qwen3 235B - Mid-size: Llama 3.3 70B (default), Qwen3 Next 80B, Mistral 31 24B - Fast: Llama 3.2 3B, Qwen3 4B - Auto-discovery from Morpheus API with fallback to static catalog - MORPHEUS_API_KEY environment variable support - Interactive onboarding via 'morpheus-api-key' auth choice - Provider auto-registration when API key is detected - Comprehensive documentation covering: - All models with context windows and features - Streaming and function calling support - Security guidelines - Troubleshooting Default model: morpheus/llama-3.3-70b (reliable, balanced performance) Morpheus API: https://api.mor.org/api/v1 (OpenAI-compatible)
2026-01-26 10:34:45 -05:00 · 2026-01-26 10:34:45 -05:00 · fa4c1245a4
commit fa4c1245a4
parent 640c8d1554
11 changed files with 725 additions and 1 deletions
--- a/docs/providers/index.md
+++ b/docs/providers/index.md
@ -45,6 +45,7 @@ See [Venice AI](/providers/venice).
 - [GLM models](/providers/glm)
 - [MiniMax](/providers/minimax)
 - [Venius (Venice AI, privacy-focused)](/providers/venice)
+- [Morpheus (decentralized inference, FREE beta)](/providers/morpheus)
 - [Ollama (local models)](/providers/ollama)

 ## Transcription providers
--- a/docs/providers/morpheus.md
+++ b/docs/providers/morpheus.md
@ -0,0 +1,222 @@
+---
+summary: "Use Morpheus decentralized inference in Clawdbot"
+read_when:
+  - You want decentralized AI inference in Clawdbot
+  - You want Morpheus API setup guidance
+---
+# Morpheus Inference API
+
+**Morpheus** provides decentralized AI inference via the Morpheus Network, offering FREE access to open-source models during Open Beta.
+
+The Morpheus Inference API is a simple, OpenAI-compatible gateway providing users access to the Morpheus Inference Marketplace. Providers host hardware and offer inference, while the API abstracts these efforts for a seamless experience.
+
+## Why Morpheus in Clawdbot
+
+- **Decentralized inference** from the Morpheus Inference Marketplace
+- **FREE during Open Beta** (until 1/31/26)
+- **20+ models** including Llama, Qwen, DeepSeek, GLM, Kimi, and more
+- OpenAI-compatible `/v1` endpoints
+
+## Features
+
+- **OpenAI-compatible API**: Standard `/v1` endpoints for easy integration
+- **Streaming**: Supported on all models
+- **Function calling**: Supported on most models
+- **Vision**: Supported on select models (e.g., `mistral-31-24b`)
+- **Web search**: Add `:web` suffix to any model for web search capabilities
+
+## Setup
+
+### 1. Get API Key
+
+1. Create an account at [app.mor.org](https://app.mor.org)
+2. Click **Create API Key** and copy it immediately
+3. Your API key format: `sk-xxxxxxxxxxxxx`
+
+### 2. Configure Clawdbot
+
+**Option A: Environment Variable**
+
+```bash
+export MORPHEUS_API_KEY="sk-xxxxxxxxxxxxx"
+```
+
+**Option B: Interactive Setup (Recommended)**
+
+```bash
+clawdbot onboard --auth-choice morpheus-api-key
+```
+
+This will:
+1. Prompt for your API key (or use existing `MORPHEUS_API_KEY`)
+2. Show all available Morpheus models
+3. Let you pick your default model
+4. Configure the provider automatically
+
+**Option C: Non-interactive**
+
+```bash
+clawdbot onboard --non-interactive \
+  --auth-choice morpheus-api-key \
+  --morpheus-api-key "sk-xxxxxxxxxxxxx"
+```
+
+### 3. Verify Setup
+
+```bash
+clawdbot chat --model morpheus/llama-3.3-70b "Hello, are you working?"
+```
+
+## Model Selection
+
+After setup, Clawdbot shows all available Morpheus models. Pick based on your needs:
+
+- **Default (our pick)**: `morpheus/llama-3.3-70b` for reliable, balanced performance
+- **Best for coding**: `morpheus/qwen3-coder-480b-a35b-instruct` with 256K context
+- **Best for reasoning**: `morpheus/kimi-k2-thinking` for deep analysis
+- **Fastest**: `morpheus/llama-3.2-3b` for low-latency responses
+
+Change your default model anytime:
+
+```bash
+clawdbot models set morpheus/llama-3.3-70b
+clawdbot models set morpheus/kimi-k2-thinking
+```
+
+List all available models:
+
+```bash
+clawdbot models list | grep morpheus
+```
+
+## Available Models
+
+### Flagship Models
+
+| Model ID | Name | Context | Best For |
+|----------|------|---------|----------|
+| `qwen3-coder-480b-a35b-instruct` | Qwen3 Coder 480B | 256K | Code generation |
+| `hermes-3-llama-3.1-405b` | Hermes 3 Llama 405B | 128K | General purpose |
+| `gpt-oss-120b` | GPT OSS 120B | 128K | GPT-style responses |
+
+### Reasoning Models
+
+| Model ID | Name | Context | Best For |
+|----------|------|---------|----------|
+| `kimi-k2-thinking` | Kimi K2 Thinking | 256K | Deep reasoning, math, coding |
+| `glm-4.7-thinking` | GLM 4.7 Thinking | 198K | Extended thinking |
+| `glm-4.7` | GLM 4.7 | 198K | Reasoning, multilingual |
+| `qwen3-235b` | Qwen3 235B | 128K | Complex reasoning |
+
+### Mid-Size Models
+
+| Model ID | Name | Context | Best For |
+|----------|------|---------|----------|
+| `llama-3.3-70b` | Llama 3.3 70B | 128K | General purpose |
+| `qwen3-next-80b` | Qwen3 Next 80B | 256K | Long context |
+| `mistral-31-24b` | Mistral 31 24B | 128K | Fast, vision |
+| `venice-uncensored` | Venice Uncensored | 32K | Uncensored, creative |
+| `hermes-4-14b` | Hermes 4 14B | 128K | Efficient |
+
+### Fast Models
+
+| Model ID | Name | Context | Best For |
+|----------|------|---------|----------|
+| `llama-3.2-3b` | Llama 3.2 3B | 128K | Fastest responses |
+| `qwen3-4b` | Qwen3 4B | 32K | Lightweight, reasoning |
+
+### Web-Enabled Models
+
+Add `:web` suffix to any model for web search:
+- `llama-3.3-70b:web`
+- `kimi-k2-thinking:web`
+- `qwen3-coder-480b-a35b-instruct:web`
+
+## Model Discovery
+
+Clawdbot automatically discovers models from the Morpheus API when `MORPHEUS_API_KEY` is set. If the API is unreachable, it falls back to a static catalog.
+
+## Streaming & Tool Support
+
+| Feature | Support |
+|---------|---------|
+| **Streaming** | All models |
+| **Function calling** | Most models |
+| **Vision/Images** | `mistral-31-24b` |
+| **JSON mode** | Supported via `response_format` |
+
+## Pricing
+
+Morpheus is **FREE during Open Beta** (until 1/31/26). Billing infrastructure will be implemented soon.
+
+## Config File Example
+
+```json5
+{
+  env: { MORPHEUS_API_KEY: "sk-..." },
+  agents: { defaults: { model: { primary: "morpheus/llama-3.3-70b" } } },
+  models: {
+    mode: "merge",
+    providers: {
+      morpheus: {
+        baseUrl: "https://api.mor.org/api/v1",
+        apiKey: "${MORPHEUS_API_KEY}",
+        api: "openai-completions",
+        models: [
+          {
+            id: "llama-3.3-70b",
+            name: "Llama 3.3 70B",
+            reasoning: false,
+            input: ["text"],
+            cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
+            contextWindow: 128000,
+            maxTokens: 8192
+          }
+        ]
+      }
+    }
+  }
+}
+```
+
+## Usage Examples
+
+```bash
+# Use default model
+clawdbot chat --model morpheus/llama-3.3-70b
+
+# Use coding model
+clawdbot chat --model morpheus/qwen3-coder-480b-a35b-instruct
+
+# Use reasoning model
+clawdbot chat --model morpheus/kimi-k2-thinking
+
+# Use with web search
+clawdbot chat --model morpheus/llama-3.3-70b:web
+```
+
+## Troubleshooting
+
+### API key not recognized
+
+```bash
+echo $MORPHEUS_API_KEY
+clawdbot models list | grep morpheus
+```
+
+Ensure the key starts with `sk-`.
+
+### Model not available
+
+Model availability depends on active providers in the Morpheus marketplace. Run `clawdbot models list` to see currently available models.
+
+### Connection issues
+
+Morpheus API is at `https://api.mor.org/api/v1`. Ensure your network allows HTTPS connections.
+
+## Links
+
+- [Morpheus API Docs](https://apidocs.mor.org)
+- [Morpheus App](https://app.mor.org)
+- [Morpheus Discord](https://discord.gg/kyVaxTHnvB)
+- [Morpheus Twitter](https://x.com/morpheusais)
--- a/src/agents/model-auth.ts
+++ b/src/agents/model-auth.ts
@ -283,6 +283,7 @@ export function resolveEnvApiKey(provider: string): EnvApiKeyResult | null {
    minimax: "MINIMAX_API_KEY",
    synthetic: "SYNTHETIC_API_KEY",
    venice: "VENICE_API_KEY",
+    morpheus: "MORPHEUS_API_KEY",
    mistral: "MISTRAL_API_KEY",
    opencode: "OPENCODE_API_KEY",
  };
--- a/src/agents/models-config.providers.ts
+++ b/src/agents/models-config.providers.ts
@ -13,6 +13,7 @@ import {
  SYNTHETIC_MODEL_CATALOG,
 } from "./synthetic-models.js";
 import { discoverVeniceModels, VENICE_BASE_URL } from "./venice-models.js";
+import { discoverMorpheusModels, MORPHEUS_BASE_URL } from "./morpheus-models.js";

 type ModelsConfig = NonNullable<MoltbotConfig["models"]>;
 export type ProviderConfig = NonNullable<ModelsConfig["providers"]>[string];
@ -350,6 +351,15 @@ async function buildVeniceProvider(): Promise<ProviderConfig> {
  };
 }

+async function buildMorpheusProvider(apiKey?: string): Promise<ProviderConfig> {
+  const models = await discoverMorpheusModels(apiKey);
+  return {
+    baseUrl: MORPHEUS_BASE_URL,
+    api: "openai-completions",
+    models,
+  };
+}
+
 async function buildOllamaProvider(): Promise<ProviderConfig> {
  const models = await discoverOllamaModels();
  return {
@ -402,6 +412,13 @@ export async function resolveImplicitProviders(params: {
    providers.venice = { ...(await buildVeniceProvider()), apiKey: veniceKey };
  }

+  const morpheusKey =
+    resolveEnvApiKeyVarName("morpheus") ??
+    resolveApiKeyFromProfiles({ provider: "morpheus", store: authStore });
+  if (morpheusKey) {
+    providers.morpheus = { ...(await buildMorpheusProvider(morpheusKey)), apiKey: morpheusKey };
+  }
+
  const qwenProfiles = listProfilesForProvider(authStore, "qwen-portal");
  if (qwenProfiles.length > 0) {
    providers["qwen-portal"] = {
--- a/src/agents/morpheus-models.ts
+++ b/src/agents/morpheus-models.ts
@ -0,0 +1,310 @@
+import type { ModelDefinitionConfig } from "../config/types.js";
+
+export const MORPHEUS_BASE_URL = "https://api.mor.org/api/v1";
+export const MORPHEUS_DEFAULT_MODEL_ID = "llama-3.3-70b";
+export const MORPHEUS_DEFAULT_MODEL_REF = `morpheus/${MORPHEUS_DEFAULT_MODEL_ID}`;
+
+// Morpheus is currently FREE during Open Beta (until 1/31/26).
+// Set costs to 0 as pricing will be implemented later.
+export const MORPHEUS_DEFAULT_COST = {
+  input: 0,
+  output: 0,
+  cacheRead: 0,
+  cacheWrite: 0,
+};
+
+/**
+ * Complete catalog of Morpheus Inference API models.
+ *
+ * Morpheus is a decentralized inference marketplace that provides access to
+ * open-source AI models. The API is fully OpenAI-compatible.
+ *
+ * Model availability depends on active providers in the marketplace.
+ * This catalog serves as a fallback when the Morpheus API is unreachable.
+ *
+ * Models with the `:web` suffix have web search capabilities enabled.
+ */
+export const MORPHEUS_MODEL_CATALOG = [
+  // ============================================
+  // FLAGSHIP MODELS
+  // ============================================
+  {
+    id: "qwen3-coder-480b-a35b-instruct",
+    name: "Qwen3 Coder 480B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 256000,
+    maxTokens: 8192,
+    tags: ["Code"],
+  },
+  {
+    id: "hermes-3-llama-3.1-405b",
+    name: "Hermes 3 Llama 3.1 405B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 128000,
+    maxTokens: 8192,
+    tags: ["General"],
+  },
+  {
+    id: "gpt-oss-120b",
+    name: "GPT OSS 120B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 128000,
+    maxTokens: 8192,
+    tags: ["General"],
+  },
+
+  // ============================================
+  // REASONING MODELS
+  // ============================================
+  {
+    id: "kimi-k2-thinking",
+    name: "Kimi K2 Thinking",
+    reasoning: true,
+    input: ["text"],
+    contextWindow: 256000,
+    maxTokens: 8192,
+    tags: ["Reasoning", "Code"],
+  },
+  {
+    id: "glm-4.7-thinking",
+    name: "GLM 4.7 Thinking",
+    reasoning: true,
+    input: ["text"],
+    contextWindow: 198000,
+    maxTokens: 8192,
+    tags: ["Reasoning"],
+  },
+  {
+    id: "glm-4.7",
+    name: "GLM 4.7",
+    reasoning: true,
+    input: ["text"],
+    contextWindow: 198000,
+    maxTokens: 8192,
+    tags: ["Reasoning"],
+  },
+  {
+    id: "qwen3-235b",
+    name: "Qwen3 235B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 128000,
+    maxTokens: 8192,
+    tags: ["Reasoning"],
+  },
+
+  // ============================================
+  // MID-SIZE MODELS
+  // ============================================
+  {
+    id: "llama-3.3-70b",
+    name: "Llama 3.3 70B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 128000,
+    maxTokens: 8192,
+    tags: ["General"],
+  },
+  {
+    id: "qwen3-next-80b",
+    name: "Qwen3 Next 80B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 256000,
+    maxTokens: 8192,
+    tags: ["General"],
+  },
+  {
+    id: "mistral-31-24b",
+    name: "Mistral 31 24B",
+    reasoning: false,
+    input: ["text", "image"],
+    contextWindow: 128000,
+    maxTokens: 8192,
+    tags: ["Vision"],
+  },
+  {
+    id: "venice-uncensored",
+    name: "Venice Uncensored",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 32000,
+    maxTokens: 8192,
+    tags: ["Uncensored"],
+  },
+  {
+    id: "hermes-4-14b",
+    name: "Hermes 4 14B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 128000,
+    maxTokens: 8192,
+    tags: ["General"],
+  },
+
+  // ============================================
+  // FAST MODELS
+  // ============================================
+  {
+    id: "llama-3.2-3b",
+    name: "Llama 3.2 3B",
+    reasoning: false,
+    input: ["text"],
+    contextWindow: 128000,
+    maxTokens: 8192,
+    tags: ["Fast"],
+  },
+  {
+    id: "qwen3-4b",
+    name: "Qwen3 4B",
+    reasoning: true,
+    input: ["text"],
+    contextWindow: 32000,
+    maxTokens: 8192,
+    tags: ["Fast", "Reasoning"],
+  },
+] as const;
+
+export type MorpheusCatalogEntry = (typeof MORPHEUS_MODEL_CATALOG)[number];
+
+/**
+ * Build a ModelDefinitionConfig from a Morpheus catalog entry.
+ */
+export function buildMorpheusModelDefinition(entry: MorpheusCatalogEntry): ModelDefinitionConfig {
+  return {
+    id: entry.id,
+    name: entry.name,
+    reasoning: entry.reasoning,
+    input: [...entry.input],
+    cost: MORPHEUS_DEFAULT_COST,
+    contextWindow: entry.contextWindow,
+    maxTokens: entry.maxTokens,
+  };
+}
+
+// Morpheus API response types
+interface MorpheusModel {
+  id: string;
+  blockchainID: string;
+  created: number;
+  tags: string[];
+  modelType: "LLM" | "TTS" | "STT" | "EMBEDDING";
+}
+
+interface MorpheusModelsResponse {
+  object: string;
+  data: MorpheusModel[];
+}
+
+/**
+ * Infer model properties from Morpheus API model data.
+ */
+function inferModelProperties(model: MorpheusModel): {
+  reasoning: boolean;
+  input: string[];
+  contextWindow: number;
+} {
+  const id = model.id.toLowerCase();
+  const tags = model.tags.map((t) => t.toLowerCase());
+
+  // Infer reasoning from model name or tags
+  const reasoning =
+    id.includes("thinking") ||
+    id.includes("reason") ||
+    id.includes("r1") ||
+    tags.includes("reasoning");
+
+  // Infer vision support from tags
+  const hasVision = id.includes("vision") || id.includes("vl-") || id.includes("-vl");
+  const input = hasVision ? ["text", "image"] : ["text"];
+
+  // Infer context window from size tag or model name
+  let contextWindow = 128000;
+  if (id.includes("qwen3-coder") || id.includes("kimi-k2") || id.includes("qwen3-next")) {
+    contextWindow = 256000;
+  } else if (id.includes("glm-4.7")) {
+    contextWindow = 198000;
+  } else if (id.includes("venice-uncensored") || id.includes("qwen3-4b")) {
+    contextWindow = 32000;
+  }
+
+  return { reasoning, input, contextWindow };
+}
+
+/**
+ * Discover models from Morpheus API with fallback to static catalog.
+ * The /models endpoint requires authentication.
+ */
+export async function discoverMorpheusModels(apiKey?: string): Promise<ModelDefinitionConfig[]> {
+  // Skip API discovery in test environment
+  if (process.env.NODE_ENV === "test" || process.env.VITEST) {
+    return MORPHEUS_MODEL_CATALOG.map(buildMorpheusModelDefinition);
+  }
+
+  // If no API key provided, use static catalog
+  if (!apiKey?.trim()) {
+    return MORPHEUS_MODEL_CATALOG.map(buildMorpheusModelDefinition);
+  }
+
+  try {
+    const response = await fetch(`${MORPHEUS_BASE_URL}/models`, {
+      headers: {
+        Authorization: `Bearer ${apiKey}`,
+      },
+      signal: AbortSignal.timeout(5000),
+    });
+
+    if (!response.ok) {
+      console.warn(
+        `[morpheus-models] Failed to discover models: HTTP ${response.status}, using static catalog`,
+      );
+      return MORPHEUS_MODEL_CATALOG.map(buildMorpheusModelDefinition);
+    }
+
+    const data = (await response.json()) as MorpheusModelsResponse;
+    if (!Array.isArray(data.data) || data.data.length === 0) {
+      console.warn("[morpheus-models] No models found from API, using static catalog");
+      return MORPHEUS_MODEL_CATALOG.map(buildMorpheusModelDefinition);
+    }
+
+    // Filter to LLM models only and merge with catalog metadata
+    const llmModels = data.data.filter((m) => m.modelType === "LLM");
+    const catalogById = new Map<string, MorpheusCatalogEntry>(
+      MORPHEUS_MODEL_CATALOG.map((m) => [m.id, m]),
+    );
+    const models: ModelDefinitionConfig[] = [];
+
+    for (const apiModel of llmModels) {
+      const catalogEntry = catalogById.get(apiModel.id);
+      if (catalogEntry) {
+        // Use catalog metadata for known models
+        models.push(buildMorpheusModelDefinition(catalogEntry));
+      } else {
+        // Create definition for newly discovered models not in catalog
+        const inferred = inferModelProperties(apiModel);
+        const displayName = apiModel.id
+          .split("-")
+          .map((part) => part.charAt(0).toUpperCase() + part.slice(1))
+          .join(" ");
+
+        models.push({
+          id: apiModel.id,
+          name: displayName,
+          reasoning: inferred.reasoning,
+          input: inferred.input as ("text" | "image")[],
+          cost: MORPHEUS_DEFAULT_COST,
+          contextWindow: inferred.contextWindow,
+          maxTokens: 8192,
+        });
+      }
+    }
+
+    return models.length > 0 ? models : MORPHEUS_MODEL_CATALOG.map(buildMorpheusModelDefinition);
+  } catch (error) {
+    console.warn(`[morpheus-models] Discovery failed: ${String(error)}, using static catalog`);
+    return MORPHEUS_MODEL_CATALOG.map(buildMorpheusModelDefinition);
+  }
+}
--- a/src/commands/auth-choice-options.ts
+++ b/src/commands/auth-choice-options.ts
@ -20,6 +20,7 @@ export type AuthChoiceGroupId =
  | "minimax"
  | "synthetic"
  | "venice"
+  | "morpheus"
  | "qwen";

 export type AuthChoiceGroup = {
@ -71,6 +72,12 @@ const AUTH_CHOICE_GROUP_DEFS: {
    hint: "Privacy-focused (uncensored models)",
    choices: ["venice-api-key"],
  },
+  {
+    value: "morpheus",
+    label: "Morpheus",
+    hint: "Decentralized inference (FREE beta)",
+    choices: ["morpheus-api-key"],
+  },
  {
    value: "google",
    label: "Google",
@ -147,6 +154,11 @@ export function buildAuthChoiceOptions(params: {
    label: "Venice AI API key",
    hint: "Privacy-focused inference (uncensored models)",
  });
+  options.push({
+    value: "morpheus-api-key",
+    label: "Morpheus API key",
+    hint: "Decentralized inference (FREE during beta)",
+  });
  options.push({
    value: "github-copilot",
    label: "GitHub Copilot (GitHub device login)",
--- a/src/commands/auth-choice.apply.api-providers.ts
+++ b/src/commands/auth-choice.apply.api-providers.ts
@ -25,6 +25,8 @@ import {
  applySyntheticProviderConfig,
  applyVeniceConfig,
  applyVeniceProviderConfig,
+  applyMorpheusConfig,
+  applyMorpheusProviderConfig,
  applyVercelAiGatewayConfig,
  applyVercelAiGatewayProviderConfig,
  applyZaiConfig,
@ -33,6 +35,7 @@ import {
  OPENROUTER_DEFAULT_MODEL_REF,
  SYNTHETIC_DEFAULT_MODEL_REF,
  VENICE_DEFAULT_MODEL_REF,
+  MORPHEUS_DEFAULT_MODEL_REF,
  VERCEL_AI_GATEWAY_DEFAULT_MODEL_REF,
  setGeminiApiKey,
  setKimiCodeApiKey,
@ -41,6 +44,7 @@ import {
  setOpenrouterApiKey,
  setSyntheticApiKey,
  setVeniceApiKey,
+  setMorpheusApiKey,
  setVercelAiGatewayApiKey,
  setZaiApiKey,
  ZAI_DEFAULT_MODEL_REF,
@ -83,6 +87,8 @@ export async function applyAuthChoiceApiProviders(
      authChoice = "synthetic-api-key";
    } else if (params.opts.tokenProvider === "venice") {
      authChoice = "venice-api-key";
+    } else if (params.opts.tokenProvider === "morpheus") {
+      authChoice = "morpheus-api-key";
    } else if (params.opts.tokenProvider === "opencode") {
      authChoice = "opencode-zen";
    }
@ -522,6 +528,65 @@ export async function applyAuthChoiceApiProviders(
    return { config: nextConfig, agentModelOverride };
  }

+  if (authChoice === "morpheus-api-key") {
+    let hasCredential = false;
+
+    if (!hasCredential && params.opts?.token && params.opts?.tokenProvider === "morpheus") {
+      await setMorpheusApiKey(normalizeApiKeyInput(params.opts.token), params.agentDir);
+      hasCredential = true;
+    }
+
+    if (!hasCredential) {
+      await params.prompter.note(
+        [
+          "Morpheus provides decentralized AI inference via the Morpheus Network.",
+          "Get your API key at: https://app.mor.org",
+          "Currently FREE during Open Beta (until 1/31/26).",
+        ].join("\n"),
+        "Morpheus Inference",
+      );
+    }
+
+    const envKey = resolveEnvApiKey("morpheus");
+    if (envKey) {
+      const useExisting = await params.prompter.confirm({
+        message: `Use existing MORPHEUS_API_KEY (${envKey.source}, ${formatApiKeyPreview(envKey.apiKey)})?`,
+        initialValue: true,
+      });
+      if (useExisting) {
+        await setMorpheusApiKey(envKey.apiKey, params.agentDir);
+        hasCredential = true;
+      }
+    }
+    if (!hasCredential) {
+      const key = await params.prompter.text({
+        message: "Enter Morpheus API key",
+        validate: validateApiKeyInput,
+      });
+      await setMorpheusApiKey(normalizeApiKeyInput(String(key)), params.agentDir);
+    }
+    nextConfig = applyAuthProfileConfig(nextConfig, {
+      profileId: "morpheus:default",
+      provider: "morpheus",
+      mode: "api_key",
+    });
+    {
+      const applied = await applyDefaultModelChoice({
+        config: nextConfig,
+        setDefaultModel: params.setDefaultModel,
+        defaultModel: MORPHEUS_DEFAULT_MODEL_REF,
+        applyDefaultConfig: applyMorpheusConfig,
+        applyProviderConfig: applyMorpheusProviderConfig,
+        noteDefault: MORPHEUS_DEFAULT_MODEL_REF,
+        noteAgentModel,
+        prompter: params.prompter,
+      });
+      nextConfig = applied.config;
+      agentModelOverride = applied.agentModelOverride ?? agentModelOverride;
+    }
+    return { config: nextConfig, agentModelOverride };
+  }
+
  if (authChoice === "opencode-zen") {
    let hasCredential = false;
    if (!hasCredential && params.opts?.token && params.opts?.tokenProvider === "opencode") {
--- a/src/commands/onboard-auth.config-core.ts
+++ b/src/commands/onboard-auth.config-core.ts
@ -10,6 +10,12 @@ import {
  VENICE_DEFAULT_MODEL_REF,
  VENICE_MODEL_CATALOG,
 } from "../agents/venice-models.js";
+import {
+  buildMorpheusModelDefinition,
+  MORPHEUS_BASE_URL,
+  MORPHEUS_DEFAULT_MODEL_REF,
+  MORPHEUS_MODEL_CATALOG,
+} from "../agents/morpheus-models.js";
 import type { MoltbotConfig } from "../config/config.js";
 import {
  OPENROUTER_DEFAULT_MODEL_REF,
@ -411,6 +417,75 @@ export function applyVeniceConfig(cfg: MoltbotConfig): MoltbotConfig {
  };
 }

+export function applyMorpheusProviderConfig(cfg: MoltbotConfig): MoltbotConfig {
+  const models = { ...cfg.agents?.defaults?.models };
+  models[MORPHEUS_DEFAULT_MODEL_REF] = {
+    ...models[MORPHEUS_DEFAULT_MODEL_REF],
+    alias: models[MORPHEUS_DEFAULT_MODEL_REF]?.alias ?? "Llama 3.3 70B",
+  };
+
+  const providers = { ...cfg.models?.providers };
+  const existingProvider = providers.morpheus;
+  const existingModels = Array.isArray(existingProvider?.models) ? existingProvider.models : [];
+  const morpheusModels = MORPHEUS_MODEL_CATALOG.map(buildMorpheusModelDefinition);
+  const mergedModels = [
+    ...existingModels,
+    ...morpheusModels.filter(
+      (model) => !existingModels.some((existing) => existing.id === model.id),
+    ),
+  ];
+  const { apiKey: existingApiKey, ...existingProviderRest } = (existingProvider ?? {}) as Record<
+    string,
+    unknown
+  > as { apiKey?: string };
+  const resolvedApiKey = typeof existingApiKey === "string" ? existingApiKey : undefined;
+  const normalizedApiKey = resolvedApiKey?.trim();
+  providers.morpheus = {
+    ...existingProviderRest,
+    baseUrl: MORPHEUS_BASE_URL,
+    api: "openai-completions",
+    ...(normalizedApiKey ? { apiKey: normalizedApiKey } : {}),
+    models: mergedModels.length > 0 ? mergedModels : morpheusModels,
+  };
+
+  return {
+    ...cfg,
+    agents: {
+      ...cfg.agents,
+      defaults: {
+        ...cfg.agents?.defaults,
+        models,
+      },
+    },
+    models: {
+      mode: cfg.models?.mode ?? "merge",
+      providers,
+    },
+  };
+}
+
+export function applyMorpheusConfig(cfg: MoltbotConfig): MoltbotConfig {
+  const next = applyMorpheusProviderConfig(cfg);
+  const existingModel = next.agents?.defaults?.model;
+  return {
+    ...next,
+    agents: {
+      ...next.agents,
+      defaults: {
+        ...next.agents?.defaults,
+        model: {
+          ...(existingModel && "fallbacks" in (existingModel as Record<string, unknown>)
+            ? {
+                fallbacks: (existingModel as { fallbacks?: string[] }).fallbacks,
+              }
+            : undefined),
+          primary: MORPHEUS_DEFAULT_MODEL_REF,
+        },
+      },
+    },
+  };
+}
+
 export function applyAuthProfileConfig(
  cfg: MoltbotConfig,
  params: {
--- a/src/commands/onboard-auth.credentials.ts
+++ b/src/commands/onboard-auth.credentials.ts
@ -100,7 +100,6 @@ export async function setSyntheticApiKey(key: string, agentDir?: string) {
 }

 export async function setVeniceApiKey(key: string, agentDir?: string) {
-  // Write to resolved agent dir so gateway finds credentials on startup.
  upsertAuthProfile({
    profileId: "venice:default",
    credential: {
@ -112,6 +111,20 @@ export async function setVeniceApiKey(key: string, agentDir?: string) {
  });
 }

+export async function setMorpheusApiKey(key: string, agentDir?: string) {
+  upsertAuthProfile({
+    profileId: "morpheus:default",
+    credential: {
+      type: "api_key",
+      provider: "morpheus",
+      key,
+    },
+    agentDir: resolveAuthAgentDir(agentDir),
+  });
+}
+
+export const MORPHEUS_DEFAULT_MODEL_REF = "morpheus/llama-3.3-70b";
+
 export const ZAI_DEFAULT_MODEL_REF = "zai/glm-4.7";
 export const OPENROUTER_DEFAULT_MODEL_REF = "openrouter/auto";
 export const VERCEL_AI_GATEWAY_DEFAULT_MODEL_REF = "vercel-ai-gateway/anthropic/claude-opus-4.5";
--- a/src/commands/onboard-auth.ts
+++ b/src/commands/onboard-auth.ts
@ -3,6 +3,10 @@ export {
  SYNTHETIC_DEFAULT_MODEL_REF,
 } from "../agents/synthetic-models.js";
 export { VENICE_DEFAULT_MODEL_ID, VENICE_DEFAULT_MODEL_REF } from "../agents/venice-models.js";
+export {
+  MORPHEUS_DEFAULT_MODEL_ID,
+  MORPHEUS_DEFAULT_MODEL_REF,
+} from "../agents/morpheus-models.js";
 export {
  applyAuthProfileConfig,
  applyKimiCodeConfig,
@ -15,6 +19,8 @@ export {
  applySyntheticProviderConfig,
  applyVeniceConfig,
  applyVeniceProviderConfig,
+  applyMorpheusConfig,
+  applyMorpheusProviderConfig,
  applyVercelAiGatewayConfig,
  applyVercelAiGatewayProviderConfig,
  applyZaiConfig,
@ -43,6 +49,7 @@ export {
  setOpenrouterApiKey,
  setSyntheticApiKey,
  setVeniceApiKey,
+  setMorpheusApiKey,
  setVercelAiGatewayApiKey,
  setZaiApiKey,
  writeOAuthCredentials,
--- a/src/commands/onboard-types.ts
+++ b/src/commands/onboard-types.ts
@ -17,6 +17,7 @@ export type AuthChoice =
  | "kimi-code-api-key"
  | "synthetic-api-key"
  | "venice-api-key"
+  | "morpheus-api-key"
  | "codex-cli"
  | "apiKey"
  | "gemini-api-key"