Frontier models, below list price.
One OpenAI- and Anthropic-compatible API for 29 models: Claude, GPT-6, Gemini 4, DeepSeek, Kimi and more. Pay per token from prepaid credits.
Sign in with GitHub, Google, Discord, Apple or an email link.
29 models, all below list
Per million tokens. You pay the first price; list is struck through. Up to 80% off. Cached input is billed at the lower cache-read rate.
| Model | Context | Input | Output | Cached input | Below list |
|---|---|---|---|---|---|
Claude Fable 5.2claude-fable-5.21M contextCached input $0.11 | 1M | $4.40 | $22.00 | $0.11 | 45% off |
Claude Opus 5.5claude-opus-5.51M contextCached input $0.11 | 1M | $2.20 | $11.00 | $0.11 | 45% off |
Claude Sonnet 5.5claude-sonnet-5.51M contextCached input $0.08 | 1M | $1.60 | $8.00 | $0.08 | 20% off |
Claude Haiku 5.5claude-haiku-5.5100K contextCached input $0.0055 | 100K | $0.055 | $0.275 | $0.0055 | 45% off |
Claude Opus 5claude-opus-51M contextCached input $0.175 | 1M | $1.75 | $8.75 | $0.175 | 65% off |
Claude Sonnet 5claude-sonnet-51M contextCached input $0.10 | 1M | $1.00 | $5.00 | $0.10 | 50% off |
GPT-6 Astragpt-6-astra1.05M contextCached input $0.60 | 1.05M | $6.00 | $30.00 | $0.60 | 40% off |
GPT-6.1 Solgpt-6.1-sol1.05M contextCached input $0.03 | 1.05M | $0.60 | $3.00 | $0.03 | 70% off |
GPT-6.1 Terragpt-6.1-terra1.05M contextCached input $0.02 | 1.05M | $0.40 | $2.00 | $0.02 | 60% off |
GPT-6 Lunagpt-6-luna1.05M contextCached input $0.0065 | 1.05M | $0.065 | $0.325 | $0.0065 | 35% off |
GPT-OSS 120Bgpt-oss-120b128K contextCached input $0.0021 | 128K | $0.014 | $0.06 | $0.0021 | 80% off |
Gemini 4 Argongemini-4-argon1.05M contextCached input $0.025 | 1.05M | $0.50 | $2.50 | $0.025 | 75% off |
Gemini 4 Carbongemini-4-carbon1.05M contextCached input $0.02 | 1.05M | $0.40 | $2.00 | $0.02 | 80% off |
Gemini 4 Hydrogengemini-4-hydrogen1.05M contextCached input $0.015 | 1.05M | $0.30 | $1.50 | $0.015 | 80% off |
Gemini 3.8 Flashgemini-3.8-flash1.05M contextCached input $0.0165 | 1.05M | $0.165 | $0.825 | $0.0165 | 80% off |
DeepSeek V4 Prodeepseek-v4-pro1M contextCached input $0.05 | 1M | $0.18 | $0.30 | $0.05 | 80% off |
DeepSeek V4.1 Flashdeepseek-v4.1-flash1.05M contextCached input $0.0003 | 1.05M | $0.01521 | $0.06084 | $0.0003 | 70% off |
Kimi K3kimi-k31M contextCached input $0.0364 | 1M | $0.0384 | $3.20 | $0.0364 | 80% off |
Kimi K3.1 Prokimi-k3.1-pro1.05M contextCached input $0.18 | 1.05M | $0.20088 | $7.57026 | $0.18 | 10% off |
Kimi K2.7 Codekimi-k2.7-code262K contextCached input $0.032 | 262K | $0.12 | $0.70 | $0.032 | 80% off |
GLM 5.3glm-5.31M contextCached input $0.0158 | 1M | $0.0168 | $0.72 | $0.0158 | 80% off |
GLM 5.3 Flashglm-5.3-flash1.05M contextCached input $0.006 | 1.05M | $0.00798 | $0.01568 | $0.006 | 80% off |
Qwen3.8 Maxqwen3.8-max1M contextCached input $0.05 | 1M | $0.40 | $1.20 | $0.05 | 80% off |
Qwen3.8 Flashqwen3.8-flash1M contextCached input $0.0032 | 1M | $0.03 | $0.094 | $0.0032 | 80% off |
MiniMax M3minimax-m3200K contextCached input $0.027 | 200K | $0.135 | $0.54 | $0.027 | 55% off |
Grok 4.7grok-4.7500K contextCached input $0.275 | 500K | $1.10 | $3.30 | $0.275 | 45% off |
MiMo V2.6 Promimo-v2.6-pro1.05M contextCached input $0.00198 | 1.05M | $0.23925 | $0.4785 | $0.00198 | 45% off |
Mistral Large 4mistral-large-4524K contextCached input $0.049 | 524K | $0.49 | $1.47 | $0.049 | 30% off |
Nemotron 3 Ultranemotron-3-ultra1M contextCached input $0.05625 | 1M | $0.1875 | $0.9375 | $0.05625 | 70% off |
The same list as JSON: GET api.routing.run/v1/pricing
Switch in three lines
Keep your SDK and your code. Point the base URL at Routing, use your key, pick a model id.
from openai import OpenAI
client = OpenAI(base_url="https://api.routing.run/v1", api_key="rk_...")
client.chat.completions.create(model="gpt-6.1-sol", messages=[{"role": "user", "content": "Hi"}])Any OpenAI client works the same way: chat completions, responses and models.
from anthropic import Anthropic
client = Anthropic(base_url="https://api.routing.run", api_key="rk_...")
client.messages.create(model="claude-opus-5.5", max_tokens=1024, messages=[{"role": "user", "content": "Hi"}])The Anthropic SDK adds /v1/messages itself, so its base URL has no /v1.
export ANTHROPIC_BASE_URL=https://api.routing.run
export ANTHROPIC_AUTH_TOKEN=rk_...
claude --model claude-opus-5.5Claude Code, shown here, reads the Anthropic variables. Codex, OpenCode, Cline and other tools take https://api.routing.run/v1 as their OpenAI base URL.
curl https://api.routing.run/v1/chat/completions \
-H "Authorization: Bearer $ROUTING_API_KEY" \
--json '{"model": "deepseek-v4-pro", "messages": [{"role": "user", "content": "Hi"}]}'Add "stream": true to get the answer as it is written.
Pay as you go
No plans and no monthly fee. You pay the per-token prices above, from prepaid credits.
Add credits
Top up from $5. Credits you buy don't expire.
Pay per token
Each request costs its uncached input tokens at the input price plus its output tokens at the output price. A key you don't use costs nothing.
Cached input is billed at each model's lower cache-read rate, listed in the models table.
See every request
The dashboard lists every request with its model, tokens and cost, next to your balance.
Questions
Anything else: support@routing.run
Why is it cheaper than list?
We buy inference capacity below list price and pass most of the saving on. The price in the table is the price you pay per token.
Do I get the model I ask for?
Yes. Every request runs the model id you send, and the table shows the context each one accepts.
What works with it?
Anything that speaks the OpenAI or Anthropic API: the official SDKs, LangChain and similar libraries, and coding agents such as Claude Code, Codex, OpenCode and Cline. Use https://api.routing.run/v1 as the OpenAI base URL, or https://api.routing.run for Anthropic clients.
Do you store my prompts?
We keep the model, token counts and cost of each request for billing. We don't store prompt or response text.
What happens if a model is unavailable?
The API answers 503 with a short message. Retry after a moment, or send the request to another model id.
Do credits expire?
Credits you buy don't expire. The smallest top-up is $5.
How do I sign in?
With GitHub, Google, Discord or Apple, or with a sign-in link sent to your email. There is no password to set.