Use PleumRouter with Any Coding Agent

It's OpenAI-compatible — point your agent at one base_url and you're done.

PleumRouter is an OpenAI-compatible API. Any agent that supports an OpenAI Compatible provider works as-is: set the base_url to https://apirouter.pleum.ai/v1 and the key to plm_…. Below are setup snippets for popular agents.

1. Common — base_url and API key#

First sign up, then issue a key under Dashboard → API Keys (keys start with plm_). Whatever the agent, all you need is the base_url (https://apirouter.pleum.ai/v1) and the key.

environment

export OPENAI_API_BASE=https://apirouter.pleum.ai/v1
export OPENAI_API_KEY=plm_xxxxxxxxxxxxxxxx
# Some agents (OpenCode, Crush) reference PLEUM_API_KEY — same key, set both.
export PLEUM_API_KEY=plm_xxxxxxxxxxxxxxxx

2. IDE · GUI agents#

These agents let you pick OpenAI Compatible (or Custom OpenAI) as the provider in settings and paste the values below — Cline, Roo Code, Kilo Code, Cursor.

Cline · Roo Code · Kilo Code · Cursor

API Provider   →  OpenAI Compatible
Base URL       →  https://apirouter.pleum.ai/v1
API Key        →  plm_xxxxxxxxxxxxxxxx
Model ID       →  gpt-4.1

Always use the OpenAI Compatible slot. The dedicated OpenRouter entry is hardcoded to openrouter.ai and will fail. Cursor requires /v1 in the base URL and a non-empty key field. Kilo Codehas a known issue (#681) where the custom base URL isn't passed to model listing, so enter the model ID manually.

Continue.dev

Continue.dev is configured via a file. With model: AUTODETECT it pulls the model list automatically.

config.yaml

# ~/.continue/config.yaml
models:
  - name: PleumRouter
    provider: openai
    apiBase: https://apirouter.pleum.ai/v1
    apiKey: plm_xxxxxxxxxxxxxxxx
    model: AUTODETECT

Zed

Zed adds an OpenAI-compatible provider in settings. Field names can vary by version — check the Zed docs.

settings.json

// Zed settings.json
{
  "language_models": {
    "openai_compatible": {
      "PleumRouter": {
        "api_url": "https://apirouter.pleum.ai/v1",
        "available_models": [
          { "name": "gpt-4.1", "max_tokens": 128000 }
        ]
      }
    }
  }
}

3. Terminal · CLI agents#

Terminal agents mostly take an OpenAI-compatible endpoint via environment variables or a config file. Goose and OpenHands need the openai/ prefix on the model name.

Aider

aider

export OPENAI_API_BASE=https://apirouter.pleum.ai/v1
export OPENAI_API_KEY=plm_xxxxxxxxxxxxxxxx

aider --model openai/gpt-4.1

OpenCode

opencode.json

// opencode.json   (or: /connect → Other)
{
  "provider": {
    "pleum": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "PleumRouter",
      "options": {
        "baseURL": "https://apirouter.pleum.ai/v1",
        "apiKey": "{env:PLEUM_API_KEY}"
      },
      "models": { "gpt-4.1": {} }
    }
  }
}

Crush

crush.json

// crush.json
{
  "providers": {
    "pleum": {
      "type": "openai-compat",
      "base_url": "https://apirouter.pleum.ai/v1",
      "api_key": "$PLEUM_API_KEY",
      "models": [{ "id": "gpt-4.1", "name": "gpt-4.1" }]
    }
  }
}

Goose · OpenHands

environment

export OPENAI_API_BASE=https://apirouter.pleum.ai/v1
export OPENAI_API_KEY=plm_xxxxxxxxxxxxxxxx

# Goose / OpenHands: prefix the model id with openai/
#   model = openai/gpt-4.1

Codex CLI

Codex CLI connects directly via the OpenAI Responses API (/v1/responses); the Chat Completions path was removed in February 2026. Add a provider with wire_api = "responses" in ~/.codex/config.toml and set PLEUM_API_KEY. Tool calls and streaming work end-to-end when routed to an OpenAI-compatible model.

~/.codex/config.toml

# ~/.codex/config.toml
# model / model_provider are document-root keys (must be above the [table]).
model = "gpt-4.1"
model_provider = "pleum"

[model_providers.pleum]
name = "PleumRouter"
base_url = "https://apirouter.pleum.ai/v1"   # Codex appends /responses → /v1/responses
env_key = "PLEUM_API_KEY"
wire_api = "responses"                      # Codex supports only the Responses API

# then:  export PLEUM_API_KEY=plm_xxxxxxxxxxxxxxxx  &&  codex

Gemini CLI upstream has weak support for external OpenAI-compatible endpoints; you may need an OpenAI-compatible wrapper or fork.

4. Claude Code (Anthropic-compatible)#

Claude Code and Claude Agent SDK tools connect via the Anthropic-compatible /v1/messages endpoint. The official CLI reads ANTHROPIC_API_KEY first, so set the same key on both ANTHROPIC_API_KEY and ANTHROPIC_AUTH_TOKEN to be safe. Point ANTHROPIC_BASE_URL at the root (no /v1) — https://apirouter.pleum.ai — because the CLI appends /v1/messages itself (adding /v1 would produce /v1/v1/messages and fail). Tools and streaming work end-to-end when routed to an OpenAI-compatible model.

Claude Code

# ANTHROPIC_BASE_URL is the ROOT (no /v1) — the CLI appends /v1/messages itself.
export ANTHROPIC_BASE_URL=https://apirouter.pleum.ai
# The official CLI prefers ANTHROPIC_API_KEY; set both to be safe.
export ANTHROPIC_API_KEY=plm_xxxxxxxxxxxxxxxx     # ← replace with your key from Dashboard > API Keys
export ANTHROPIC_AUTH_TOKEN=plm_xxxxxxxxxxxxxxxx  # ← same key

# Pass a PleumRouter model id (see GET /v1/models for the full list):
claude --model anthropic/gpt-4.1

The plm_xxxxxxxxxxxxxxxx in the snippet is a placeholder. Replace it with a real key (starting with plm_) issued under API Keys in the dashboard. Find model IDs via GET /v1/models or on the models page.

5. Model IDs and auto-discovery#

Find model IDs at GET /v1/models or on the models page. Cline, Continue, OpenCode and others fetch this list automatically to fill their dropdowns. OpenRouter-format IDs (openai/gpt-5.5) are converted automatically.

list models

curl https://apirouter.pleum.ai/v1/models \
  -H "Authorization: Bearer plm_xxxxxxxxxxxxxxxx"

6. Image · audio · video (direct API)#

Image, audio and video are also OpenAI-compatible HTTP endpoints — image POST /v1/images/generations, TTS POST /v1/audio/speech, STT POST /v1/audio/transcriptions, and video POST /v1/video/generations (async — returns a job_id, then poll GET /v1/jobs/{job_id}). Swap MODEL_ID for a model that supports the modality — see the models page or GET /v1/models.

multimodal endpoints

# Image generation
curl https://apirouter.pleum.ai/v1/images/generations \
  -H "Authorization: Bearer plm_xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{ "model": "MODEL_ID", "prompt": "a red bicycle", "n": 1, "size": "1024x1024" }'

# Text-to-speech (returns audio bytes; cost in X-Cost-Krw header)
curl https://apirouter.pleum.ai/v1/audio/speech \
  -H "Authorization: Bearer plm_xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{ "model": "MODEL_ID", "input": "Hello there", "voice": "alloy" }' --output speech.mp3

# Speech-to-text (multipart upload)
curl https://apirouter.pleum.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer plm_xxxxxxxxxxxxxxxx" \
  -F model=MODEL_ID -F file=@audio.mp3

# Video generation is async: POST returns a job_id, then poll GET /v1/jobs/{job_id}
curl https://apirouter.pleum.ai/v1/video/generations \
  -H "Authorization: Bearer plm_xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{ "model": "MODEL_ID", "prompt": "a drone shot over a forest" }'

7. More agents#

Most agents not listed here work the same way. Via OpenAI-compatible base_url — OpenHands, Open Interpreter, SWE-agent, Qwen Code, MetaGPT, GPT-Pilot, ChatDev, Tabby (chat), Dyad, Plandex, bolt.diy, Forge, Kimi CLI, gptme, Letta, and more. Via Anthropic-compatible (/v1/messages): claude-code-router also connects with ANTHROPIC_BASE_URL. LiteLLM-based agents (Aider, OpenHands, Open Interpreter, SWE-agent, gptme) connect directly via base_url since PleumRouter is already a gateway — but if you keep a LiteLLM proxy, configure it as below (don't append /chat/completions to api_base).

litellm config.yaml

# litellm config.yaml  (LiteLLM proxy를 쓰는 경우)
model_list:
  - model_name: pleum-gpt-4.1
    litellm_params:
      model: openai/gpt-4.1              # openai/ prefix → OpenAI-compat route
      api_base: https://apirouter.pleum.ai/v1           # do NOT append /chat/completions
      api_key: os.environ/PLEUM_API_KEY

Stuck somewhere? Drop your key into the playground and test it right away. You also earn rewards for signing up, linking an account, your first call, and your first payment — up to ₩5,000.