python from together import Together client = Together(apikey="...") response = client.chat.completions.create( model="meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo", messages=[{"role": "user", "content": "Hello"}] ) print(response.choices[0].message.content) OpenAI-compatible: same shape.

python stream = client.chat.completions.create( model="meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo", messages=[{"role": "user", "content": "..."}], stream=True ) for chunk in stream: print(chunk.choices[0].delta.content or "", end="")

LLMs - meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo - mistralai/Mixtral-8x22B-Instruct-v0.1 - Qwen/Qwen2.5-72B-Instruct-Turbo - deepseek-ai/DeepSeek-V3 - google/gemma-2-27b-it Code - Qwen/Qwen2.5-Coder-32B-Instruct - meta-llama/CodeLlama-70b-Instruct-hf Image - black-forest-labs/FLUX.1-pro - stabilityai/stable-diffusion-xl Embeddings - BAAI/bge-large-en-v1.5 - sentence-transformers/all-MiniLM-L6-v2

Per million tokens: - Llama 405B: $5 / $5 - Llama 70B: $0.88 / $0.88 - Mixtral 8x22B: $1.20 / $1.20 - DeepSeek V3: $1.20 / $1.20 - Qwen 72B: $0.90 / $0.90 Compare: - GPT-5: $15 / $60 - Claude Opus: $15 / $75 Together is 5-10× cheaper.

python client.finetuning.create( trainingfile="file-...", model="meta-llama/Meta-Llama-3.1-8B-Instruct", nepochs=3, ) Cheaper than OpenAI fine-tuning. More model choice.

python response = client.images.generate( model="black-forest-labs/FLUX.1-pro", prompt="An astronaut riding a horse", ) print(response.data[0].url)

python embedding = client.embeddings.create( model="BAAI/bge-large-en-v1.5", input="The text to embed" ).data[0].embedding

Open-source preference - Cost-sensitive (5-10× cheaper than closed) - Want flexibility to switch providers (OpenAI-compatible) - Need fine-tuning at competitive cost

How to Use Together AI 2026 — Guide

Q: Why Together

Wide model selection (Llama, Mistral, Qwen, DeepSeek, etc.) - Competitive pricing (often cheaper than closed APIs) - Fast inference (optimized) - Fine-tuning service - OpenAI-compatible API

How to Use Together AI 2026

Together AI hosts open-source models for production inference. Why Together - Wide model selection Llama, Mistral, Qwen, DeepSeek, etc. - Competitive pricing often cheaper than closed APIs - Fast inference optimized - Fine-tuning service - OpenAI-compatible API Sign Up api.together.xyz → sign up. Get $5 free credit. Install bash pip install together First Request python from together import Toget…

All guides

Related guides

How to Start AI SaaS 2026
How to Use v0 by Vercel 2026 — Complete Walkthrough
How to Become a Vibe Coder: 2026 Roadmap
How to Build an AI Agent 2026
How to Build a Chatbot 2026