freetiermodels

Why This Model

Gemma 4 has generous daily rate limits, while Gemini 3.1 Flash-Lite has 500 RPD.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "google/gemini-3.1-flash-lite-latest",
  "provider": {
    "google": {
      "options": {
        "apiKey": "{env:GOOGLE_API_KEY}"
      }
    }
  }
}

Why This Model

Very generous free tier, though subject to RPM rate limits.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "mistral/mistral-small-3.1",
  "provider": {
    "mistral": {
      "options": {
        "apiKey": "{env:MISTRAL_API_KEY}"
      }
    }
  }
}

Why This Model

Good rate limits for this lesser-known model.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "cohere/command-a",
  "provider": {
    "cohere": {
      "options": {
        "apiKey": "{env:COHERE_API_KEY}"
      }
    }
  }
}

Why This Model

Long signup process but generous limits. Unsure of the current RPD limits for the free tier.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "nvidia/meta/llama-3.1-8b-instruct",
  "provider": {
    "nvidia": {
      "options": {
        "apiKey": "{env:NVIDIA_API_KEY}"
      }
    }
  }
}

Why This Model

Good starting point for free inference, I recommend topping up $10.50 to get 1,000 free requests per day (from 50).

{
  "$schema": "https://opencode.ai/config.json",
  "model": "nvidia/nemotron-3-super-120b-a12b:free",
  "provider": {
    "openrouter": {
      "options": {
        "apiKey": "{env:OPENROUTER_API_KEY}"
      }
    }
  }
}

Why This Model

Free models available for limited time while teams collect feedback. Rate limits may be hit very quickly if congested.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "opencode/minimax-m2.5-free",
  "provider": {
    "opencode": {
      "options": {
        "apiKey": "{env:OPENCODE_ZEN_API_KEY}"
      }
    }
  }
}

Why This Model

Multiple free models available with decent usage, but responses have been slow for me.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "kilo/nvidia/nemotron-3-super-120b-a12b:free",
  "provider": {
    "kilo": {
      "options": {
        "apiKey": "{env:KILO_API_KEY}"
      }
    }
  }
}

Why This Model

Good variety of models, but rate limits get used up quickly.

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "ollama": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Ollama Cloud",
      "options": {
        "baseURL": "https://ollama.com/v1"
      },
      "models": {
        "glm-4.7:cloud": {
          "name": "glm-4.7:cloud"
        }
      }
    }
  }
}

Why This Model

50 free requests per month is very limited, but we get access to Claude Haiku and GPT-5 mini

{
  "$schema": "https://opencode.ai/config.json",
  "model": "github/claude-haiku-4.5",
  "provider": {
    "github": {
      "options": {
        "token": "{env:GITHUB_TOKEN}"
      }
    }
  }
}

Why This Model

While rate limits are very generous, I constantly get 'Reasoning not supported with this model' errors when using with OpenCode.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "groq/llama-3.1-8b-instant",
  "provider": {
    "groq": {
      "options": {
        "apiKey": "{env:GROQ_API_KEY}"
      }
    }
  }
}

Why This Model

Limits nerfed for GLM-4.7 and GPT-OSS, I constantly get 'Session too large to compact - context exceeds model limit even after stripping media' errors when using with OpenCode.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "cerebras/llama-3.3-70b",
  "provider": {
    "cerebras": {
      "options": {
        "apiKey": "{env:CEREBRAS_API_KEY}"
      }
    }
  }
}

Why This Model

Rate limits get used up very quickly.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "github/gpt-4o-mini",
  "provider": {
    "github": {
      "options": {
        "token": "{env:GITHUB_TOKEN}"
      }
    }
  }
}

Why This Model

Long signup process, Cloudflare uses 'neurons' instead of credits and they get used up very quickly with the free tier.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "cloudflare/@cf/meta/llama-3.1-8b-instruct",
  "provider": {
    "cloudflare": {
      "options": {
        "accountId": "{env:CF_ACCOUNT_ID}",
        "apiToken": "{env:CF_API_TOKEN}"
      }
    }
  }
}

Why This Model

$0.10/mo in free credits is extremely limited.

{
  "$schema": "https://opencode.ai/config.json",
  "model": "huggingface/meta-llama/Llama-3.2-3B-Instruct",
  "provider": {
    "huggingface": {
      "options": {
        "token": "{env:HF_TOKEN}"
      }
    }
  }
}

{
  "$schema": "https://opencode.ai/config.json",
  "model": "kilocode/KiloCode-Claude-Sonnet",
  "provider": {
    "kilocode": {
      "options": {
        "apiKey": "{env:KILOCODE_API_KEY}"
      }
    }
  }
}

{
  "$schema": "https://opencode.ai/config.json",
  "model": "fireworks/llama-v3p2-3b-instruct",
  "provider": {
    "fireworks": {
      "options": {
        "apiKey": "{env:FIREWORKS_API_KEY}"
      }
    }
  }
}

Why This Model

Free tier with no request limits, but I had to continuously prompt it to continue.

Setup via Add Provider Skill

1. Install OpenCode
2. Install the add-provider skill
3. Run add provider and choose Crof
4. Enter your API key when prompted

{
  "$schema": "https://opencode.ai/config.json",
  "model": "crof/glm-4.7-flash",
  "provider": {
    "crof": {
      "options": {
        "apiKey": "{env:CROF_API_KEY}"
      }
    }
  }
}

Why This Model

Requires a one-time $10 top-up to access the free tier, models are powered by OpenRouter.

Setup via Add Provider Skill

1. Install OpenCode
2. Install the add-provider skill
3. Run add provider and choose Nous Portal
4. Enter your API key when prompted

{
  "$schema": "https://opencode.ai/config.json",
  "model": "nousresearch/hermes-3-llama-3.1-405b",
  "provider": {
    "openai-compatible": {
      "options": {
        "baseURL": "https://gateway.nous.chat/v1",
        "apiKey": "{env:NOUS_API_KEY}"
      }
    }
  }
}

Why This Model

Lite plan gives 400 requests/day. Requires MCP installation via opencode-add-provider skill.

Setup via Add Provider Skill

1. Install OpenCode
2. Install the add-provider skill
3. Run add provider and choose routing.run
4. Enter your API key when prompted

{
  "$schema": "https://opencode.ai/config.json",
  "model": "routing/openrouter/free",
  "provider": {
    "routing": {
      "options": {
        "apiKey": "{env:ROUTING_API_KEY}"
      }
    }
  }
}