Connect Agents

Connect to OpenCode

CheapestToken is listed in the OpenCode provider catalog like any other provider: search for it, connect your key, and every model is selectable.

Get an API Key

Create an account and a prepaid API key in the console — the quickstart walks through it step by step. Keys can be revoked at any time from the console.

Connect in the Terminal

Run /connect and find Cheapest Token in the provider list — once the models.dev listing merges it sits in the catalog exactly like DigitalOcean, searchable by name — paste your sk- key, then run /models to pick any model.

Until the listing merges: scroll to Other, enter the provider id cheapest-token, then paste the key.

/connect
# scroll to "Other", enter the provider id:
cheapest-token
# then paste your sk-… API key when prompted

Connect in OpenCode Desktop

Open Settings → the Providers tab, press View all (or use the search) and type Cheapest Token, hit Connect and paste your sk- key — every catalog model then appears in the model picker. It is the same key as the terminal app, and usage bills from your prepaid wallet.

Before the Listing Ships — Manual Config

Only if you want to start today, before the public listing is merged — afterwards the search-and-connect flow above is all you need. The blocks below are generated live from the catalog on every visit, so they always match the Models page.

OpenCode 2 / desktop

{
  "$schema": "https://opencode.ai/config.json",
  "providers": {
    "cheapest-token": {
      "name": "CheapestToken",
      "env": [
        "CHEAPEST_TOKEN_API_KEY"
      ],
      "package": "@opencode/ai/providers/openai-compatible",
      "settings": {
        "baseURL": "https://api.cheapest-token.com/v1"
      },
      "models": {
        "deepseek-v4-pro-0813": {
          "name": "DeepSeek V4 Pro 0813",
          "limit": {
            "context": 1048576
          }
        },
        "deepseek-v4.1-flash-1017": {
          "name": "DeepSeek V4.1 Flash 1017",
          "limit": {
            "context": 1048576
          }
        },
        "glm-5.3": {
          "name": "GLM5.3",
          "limit": {
            "context": 1048576
          }
        },
        "glm-5.3-flash": {
          "name": "GLM5.3 Flash",
          "limit": {
            "context": 1048576
          }
        },
        "gpt-5.5": {
          "name": "GPT-5.5",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-5.6-luna": {
          "name": "GPT-5.6 Luna",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-5.6-sol": {
          "name": "GPT-5.6 Sol",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-5.6-terra": {
          "name": "GPT-5.6 Terra",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6-astra": {
          "name": "GPT-6 Astra",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6-luna": {
          "name": "GPT-6 Luna",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6-sol": {
          "name": "GPT-6 Sol",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6.1-sol": {
          "name": "GPT-6.1 Sol",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6.1-terra": {
          "name": "GPT-6.1 Terra",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-chat-latest": {
          "name": "GPT-Chat Latest",
          "limit": {
            "context": 400000
          }
        },
        "grok-4.6": {
          "name": "Grok 4.6",
          "limit": {
            "context": 500000
          }
        },
        "grok-4.7": {
          "name": "Grok 4.7",
          "limit": {
            "context": 500000
          }
        },
        "kimi-k3": {
          "name": "Kimi K3",
          "limit": {
            "context": 1048576
          }
        },
        "kimi-k3.1-pro": {
          "name": "Kimi K3.1 Pro",
          "limit": {
            "context": 1048576
          }
        },
        "minimax-m3.1-flash": {
          "name": "MiniMax M3.1-Flash",
          "limit": {
            "context": 1000000
          }
        },
        "nemotron-3.5-super-vl": {
          "name": "NVIDIA Nemotron 3.5 Super VL",
          "limit": {
            "context": 1048576
          }
        },
        "qwen3.8-max": {
          "name": "Qwen3.8-Max",
          "limit": {
            "context": 1000000
          }
        }
      }
    }
  }
}

OpenCode 1 / terminal

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "cheapest-token": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "CheapestToken",
      "options": {
        "baseURL": "https://api.cheapest-token.com/v1"
      },
      "models": {
        "deepseek-v4-pro-0813": {
          "name": "DeepSeek V4 Pro 0813",
          "limit": {
            "context": 1048576
          }
        },
        "deepseek-v4.1-flash-1017": {
          "name": "DeepSeek V4.1 Flash 1017",
          "limit": {
            "context": 1048576
          }
        },
        "glm-5.3": {
          "name": "GLM5.3",
          "limit": {
            "context": 1048576
          }
        },
        "glm-5.3-flash": {
          "name": "GLM5.3 Flash",
          "limit": {
            "context": 1048576
          }
        },
        "gpt-5.5": {
          "name": "GPT-5.5",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-5.6-luna": {
          "name": "GPT-5.6 Luna",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-5.6-sol": {
          "name": "GPT-5.6 Sol",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-5.6-terra": {
          "name": "GPT-5.6 Terra",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6-astra": {
          "name": "GPT-6 Astra",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6-luna": {
          "name": "GPT-6 Luna",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6-sol": {
          "name": "GPT-6 Sol",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6.1-sol": {
          "name": "GPT-6.1 Sol",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-6.1-terra": {
          "name": "GPT-6.1 Terra",
          "limit": {
            "context": 1050000
          }
        },
        "gpt-chat-latest": {
          "name": "GPT-Chat Latest",
          "limit": {
            "context": 400000
          }
        },
        "grok-4.6": {
          "name": "Grok 4.6",
          "limit": {
            "context": 500000
          }
        },
        "grok-4.7": {
          "name": "Grok 4.7",
          "limit": {
            "context": 500000
          }
        },
        "kimi-k3": {
          "name": "Kimi K3",
          "limit": {
            "context": 1048576
          }
        },
        "kimi-k3.1-pro": {
          "name": "Kimi K3.1 Pro",
          "limit": {
            "context": 1048576
          }
        },
        "minimax-m3.1-flash": {
          "name": "MiniMax M3.1-Flash",
          "limit": {
            "context": 1000000
          }
        },
        "nemotron-3.5-super-vl": {
          "name": "NVIDIA Nemotron 3.5 Super VL",
          "limit": {
            "context": 1048576
          }
        },
        "qwen3.8-max": {
          "name": "Qwen3.8-Max",
          "limit": {
            "context": 1000000
          }
        }
      }
    }
  }
}

For env-based auth instead of /connect, export CHEAPEST_TOKEN_API_KEY — the v2 shape above reads it via its env key; in the v1 shape set options.apiKey to { "env": "CHEAPEST_TOKEN_API_KEY" }. limit.context is each model's full window; per-1M prices are on the Models page.

What to Expect

Every catalog model appears under the CheapestToken provider in OpenCode's /models picker.
Streaming and tool calls are supported.
Usage bills from your prepaid wallet at the prices shown on the Models page.
The catalog endpoint GET /api/v1/models (Models API) stays the source of truth — new models land in the config on your next visit.

Was this helpful?