Together¶
Together AI's platform for accessing various open-source models.
Configuration¶
PROVIDER=together
API_KEY=your-together-api-key
MODEL=meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
TEMPERATURE=0.3
MAX_TOKENS=1024
Key Parameters¶
| Parameter | Description | Default |
|---|---|---|
API_KEY |
Together API key (required) | - |
MODEL |
Model to use | - |
TEMPERATURE |
Randomness (0.0-1.0) | 0.3 |
TOP_P |
Nucleus sampling | 1.0 |
MAX_TOKENS |
Max response tokens | 1024 |
TIMEOUT |
Request timeout (seconds) | 60 |
BASE_URL |
Custom API endpoint | https://api.together.xyz/v1 |
EXTRA_BODY |
Additional request parameters | {} |
Features¶
- ✅ Streaming responses
- ✅ Function calling
- ✅ MCP support
- ✅ OpenAI-compatible API
- ✅ Multiple open-source models
- ✅ Cost-effective inference
Important Notes¶
- Uses OpenAI-compatible client implementation
- Access to various open-source models
- Competitive pricing for open-source models
- Supports latest Llama, Qwen, and other models
- Good alternative for open-source model access