Fireworks AI Provider

Fireworks AI provides fast inference for open-source models with optimized serving infrastructure and competitive pricing.

Get an API Key

1. Go to fireworks.ai
2. Create an account or sign in
3. Navigate to API Keys
4. Create a new key and copy it

Fireworks AI offers a free tier with limited usage.

Configure VibeCody

Option 1: Environment variable (recommended)

export FIREWORKS_API_KEY="..."
vibecli --provider fireworks

Option 2: Config file (~/.vibecli/config.toml)

[fireworks]
enabled = true
api_key = "..."
model = "accounts/fireworks/models/llama-v3p1-70b-instruct"

Model Selection

Default: accounts/fireworks/models/llama-v3p1-70b-instruct

Override from the CLI:

vibecli --provider fireworks --model "accounts/fireworks/models/llama-v3p1-8b-instruct"

Best For

• Fast inference – optimized serving for low-latency responses
• Open-source models – access Llama, Mixtral, and other open models
• Free tier – experiment before committing

Verify Connection

vibecli --provider fireworks -c "Write a Go HTTP handler with middleware"

Troubleshooting

Invalid API key

• Check your key at fireworks.ai
• Confirm the env var is set: echo $FIREWORKS_API_KEY

Model not found

• Use the full model path including accounts/fireworks/models/ prefix
• Check fireworks.ai/models for current model names