OpenCode Go

OpenCode Go is a low-cost subscription plan that provides reliable access to popular open coding models. It is designed for developers who want a budget-friendly option without compromising on model quality.

Overview

OpenCode Go focuses on open-weight and open-source models that are tested and verified by the opencode team. It offers a cost-effective alternative to premium subscriptions and individual provider accounts.

Key Benefits

• Low cost — Affordable subscription pricing for budget-conscious developers.
• Open models — Access to popular open-weight models like Llama, DeepSeek, Mistral, and Qwen.
• Tested and verified — Every model on Go is validated for coding performance.
• Reliable access — Hosted and maintained by opencode, no self-hosting required.
• Simple setup — One subscription, one API key.

Setup

Step 1: Connect

Open the command palette and run:

/connect

Step 2: Select OpenCode Go

From the provider list, choose OpenCode Go.

Step 3: Sign In

A browser window opens to https://opencode.ai/auth. Sign in with your opencode account. If you do not have one, create a new account.

Step 4: Add Billing Details

After signing in:

1. Navigate to the subscription section.
2. Select the OpenCode Go plan.
3. Add your billing information (credit card or other supported payment method).
4. Confirm the subscription.

Step 5: Get Your API Key

1. Go to the API Keys section of your dashboard.
2. Generate a new API key for OpenCode Go.
3. Copy the key.

Step 6: Paste the Key

Return to the terminal and paste the API key when prompted.

Step 7: Choose a Model

Run:

/models

Select from the available open models.

Available Models

OpenCode Go includes a rotating selection of open-weight models. The exact lineup changes as new models are released and evaluated. Typical categories include:

Category	Example Models	Best For
General Purpose	Llama 4, DeepSeek V3	Everyday coding, chat, explanation
Code Specialized	DeepSeek Coder, CodeLlama	Code generation, debugging, refactoring
Lightweight	Mistral 7B, Qwen 2.5	Quick edits, completions, low-latency tasks
Large Context	YaRN-based models	Long-file analysis, full-project context

Run /models in opencode to see the current list.

Model Quality

All models on OpenCode Go are tested against opencode's internal coding benchmark suite. The results are published transparently — each model listing includes:

• Pass rate on coding tasks.
• Latency — average time to first token.
• Context window — maximum supported context length.
• Strengths — types of tasks where the model excels.

Billing

Subscription Management

1. Visit https://opencode.ai/account/subscription.
2. View your current plan details.
3. Upgrade, downgrade, or cancel.

Pricing

OpenCode Go is priced lower than premium plans. Specific pricing is available at https://opencode.ai/pricing. The plan includes a monthly token allowance with the option to purchase additional tokens.

Fair Use

OpenCode Go operates under a fair-use policy:

• Reasonable daily usage is included in the subscription.
• Excessive usage may be throttled or require an upgrade.
• The dashboard shows your current billing period usage.

Why Use OpenCode Go?

Compared to Self-Hosting

Aspect	OpenCode Go	Self-Hosted Open Model
Cost	Low subscription	Hardware + electricity costs
Setup	5 minutes	Hours to days
Maintenance	None	Ongoing model updates, infrastructure
Reliability	99.9% uptime SLAs	Dependent on your hardware
Model selection	Curated, tested	Any model you can run

Compared to Premium APIs

OpenCode Go fills the gap between free limited-tier APIs and expensive premium subscriptions. It is ideal for:

• Hobbyists and independent developers.
• Teams with budget constraints.
• Prototyping and experimentation.
• Projects where open-weight models are preferred for licensing or transparency reasons.

Troubleshooting

Connection Issues

• Verify your internet connection.
• Ensure https://opencode.ai is accessible.
• Check that your API key is correctly entered.

Rate Limiting

If you encounter rate limits:

• Check your usage dashboard at https://opencode.ai/account/usage.
• Wait for the rate limit window to reset (typically 1 minute).
• Consider upgrading to a higher tier for increased limits.

Model Quality Concerns

If a model's output quality is below expectations:

• Try a different model from the Go lineup — different models excel at different tasks.
• Ensure the model's context window is appropriate for your input size.
• Report quality issues through the opencode feedback channel.

Billing Problems

• Visit the subscription management page at https://opencode.ai/account/subscription.
• Verify your payment method is up to date.
• Contact opencode support for billing disputes or questions.

Comparison: OpenCode Go vs OpenCode Zen

Feature	OpenCode Go	OpenCode Zen
Price	Low cost	Premium
Models	Open-weight models	Curated mix including proprietary
Best for	Budget-conscious, open-model preference	Best-in-class model access
Setup	Single connection	Single connection
Testing	Verified by opencode team	Verified by opencode team

Security

• API keys are sensitive credentials. Store them securely.
• Use environment variables or opencode's secure credential storage rather than hardcoding.
• Revoke and regenerate keys from the dashboard if they are compromised.
• OpenCode Go uses encrypted API communication end-to-end.