Together AI alternative.
Frontier + open-weight under one key — not just open-source.

from together import Together

client = Together(api_key="...")

r = client.chat.completions.create(
    model="meta-llama/Llama-4-Scout-17B-16E-Instruct",
    messages=[{"role": "user", "content": "Hello"}],
)

# pip install aigateway-py openai
# aigateway-py: sub-accounts, evals, replays, jobs, webhook verify.
# openai SDK: chat/embeddings/images/audio — drop-in compat per our SDK's own guidance.
from openai import OpenAI

client = OpenAI(
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)

r = client.chat.completions.create(
    model="meta/llama-4-scout-17b-16e-instruct",
    messages=[{"role": "user", "content": "Hello"}],
)

• Frontier AND open-weight on one key — Claude Opus 4.7, GPT-5.4, Gemini 3.1 Pro, Kimi K2.6, Llama 4 together.
• Every modality beyond text — image, video, voice, audio, music, embeddings.
• Eval-driven routing to pick the cheapest frontier-quality option.
• Sub-account API — mint keys per end user with spend caps.
• Edge latency — 47ms p50 overhead globally vs Together's regional cold start.

• Together can host fine-tuned custom weights we don't yet support; fine-tuning is on the Enterprise roadmap.