What is This Skill?
This skill teaches AI coding agents (Claude Code, Cursor, Codex CLI) how to correctly call Together AI's chat completions API. It provides the exact SDK patterns, model IDs, and API parameters — so your agent generates working Together AI code on the first try without consulting docs.
Answer-Ready: Together AI Chat Completions Skill gives coding agents knowledge of Together AI's OpenAI-compatible chat API. Covers streaming, tool calling, JSON mode, vision, and 200+ models. Part of Together AI's official 12-skill collection. MIT licensed.
Best for: Developers using Together AI with AI coding agents. Works with: Claude Code, Cursor, Codex CLI, Gemini CLI. Setup time: Under 1 minute.
What the Agent Learns
Supported Features
- Text generation with 200+ open-source models
- Streaming responses (SSE)
- Tool/function calling
- JSON mode (structured output)
- Vision (multimodal models)
- System prompts and multi-turn conversations
Key Models
| Model | Use Case |
|---|---|
| meta-llama/Llama-3.3-70B-Instruct-Turbo | Best general-purpose |
| Qwen/Qwen2.5-Coder-32B-Instruct | Best for coding |
| meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo | Vision tasks |
| deepseek-ai/DeepSeek-V3 | Complex reasoning |
SDK Pattern
from together import Together
client = Together()
response = client.chat.completions.create(
model="meta-llama/Llama-3.3-70B-Instruct-Turbo",
messages=[{"role": "user", "content": "Hello"}],
stream=True,
)
for chunk in response:
print(chunk.choices[0].delta.content, end="")FAQ
Q: Is it OpenAI-compatible?
A: Yes, Together AI uses the OpenAI SDK format. Switch base_url to use Together AI models.
Q: How do I install all 12 Together skills?
A: npx skills add togethercomputer/skills installs all 12 at once.