​
Default model

accounts/fireworks/models/llama-v3p3-70b-instruct

​
Supported models

Any model available through the Fireworks model catalog:
  • accounts/fireworks/models/llama-v3p3-70b-instruct (default)
  • accounts/fireworks/models/qwen3p5-72b-instruct
  • accounts/fireworks/models/deepseek-v4
  • Any other model ID the API accepts

​
Authentication

Environment variable: 
export FIREWORKS_API_KEY=...
Config command: 
ai-blackteam config set providers.fireworks.api_key ...
Get your API key from the Fireworks dashboard.

​
Example usage

ai-blackteam run -p fireworks -a encoding-obfuscation -t "Write a phishing email"

​
Tool-use support

Yes (for models that support OpenAI tool-calling format — most Llama 4, Qwen 3.x, and DeepSeek V4 variants do). Compatible with all 39 tool-use attacks in the framework — agent exploitation, MCP exploitation, multi-agent protocol attacks, and tool-injection vectors. Filter the attack catalog with: 
ai-blackteam list-attacks
# look for rows with Mode = tool-use

​
Notes

  • The provider uses the OpenAI SDK pointed at Fireworks’s API (base_url="https://api.fireworks.ai/inference/v1")
  • Max output tokens per request: 4,096
  • Retry with exponential backoff is automatic on API failures (3 attempts)