Archer is one OpenAI-compatible endpoint that routes each request to the model best suited for it. One key, no model-picking — every answer comes back as archer-auto.
A request travels like an arrow — nocked, aimed, loosed. Here is exactly what happens between your call and the answer.
You send a standard chat request — the same JSON you'd send OpenAI. Your model field is accepted, then ignored.
Archer reads the intent of your prompt and selects the model matched to it — coding, math, analysis, or a fast simple reply.
The chosen model answers. If it's rate-limited or errors, Archer falls through the chain until one lands — you never see the miss.
A curated pool sits behind your key. You never pick from it — Archer draws the right one for every request.
Groq
llama-3.3-70b
Coding, reasoning & the default shot
Groq
gpt-oss-120b
Math and hard, multi-step problems
Groq
llama-4-scout
Writing and long-context drafting
Groq
llama-3.1-8b
Fast replies to short, simple asks
Groq
gpt-oss-20b
Quick conversational turns
Ollama
qwen3-coder-480b
Coding across huge codebases
Ollama
glm-4.7
General reasoning & writing
Ollama
minimax-m3
Very long-context analysis
Ollama
nemotron-3-super
Analysis & step-by-step reasoning
Whoever answers, your response always comes back as archer-auto.
If you've called OpenAI, you've already written Archer. Point the client at our URL, use your key, and route.
from openai import OpenAI
client = OpenAI(
api_key="arch_sk_...",
base_url="https://your-archer.app/v1",
)
resp = client.chat.completions.create(
model="archer-auto", # ignored — Archer picks
messages=[{"role": "user",
"content": "Write a binary search in Rust"}],
)
print(resp.choices[0].message.content)Each prompt is read and sent to the model that fits it — code, math, analysis, or a quick reply.
Rate-limited or down? Archer retries down a fixed chain until a model lands the shot.
A drop-in /v1/chat/completions endpoint. Keep your SDK — just swap the base URL.
A single arch_sk_ key stands in for the wallet of provider keys you'd otherwise juggle.
See which model answered, why it routed there, plus tokens and latency, on your dashboard.
Whoever answers, the response always comes back in the same archer-auto shape.
Sign up, generate a key, and point your first request at Archer. The right model is already waiting on the line.