Change the base URL and the key. Model IDs such as anthropic/claude-fable-5.1 keep working. Images, video and voice use the same key. Pay with a Russian bank card, by bank transfer, or in USDT or TON.
Original models from their makers — no lighter substitutes
A $10 test balance — ask support after you sign up. Message us on Telegram
client = OpenAI(- base_url="https://openrouter.ai/api/v1",- api_key=os.environ["OPENROUTER_API_KEY"],+ base_url="https://gateway.api429.com/v1",+ api_key=os.environ["API429_API_KEY"], ) client.chat.completions.create( model="anthropic/claude-fable-5.1",Same model ID messages=messages, )OpenRouter model rates and current API429 prices for models with an exact ID match. Each cell shows input / output.
| Model | OpenRouter ID | OpenRouter | API429 | Difference |
|---|---|---|---|---|
| Claude Fable 5.1 | anthropic/claude-fable-5.1 | $10 / $50 | $3 / $15 | −70% |
| Claude Opus 5 | anthropic/claude-opus-5 | $5 / $25 | $1.50 / $7.50 | −70% |
| Claude Sonnet 5 | anthropic/claude-sonnet-5 | $2 / $10 | $0.60 / $3 | −70% |
| GPT 6.1 Sol | openai/gpt-6.1-sol | $2 / $10 | $0.60 / $3 | −70% |
| GPT 6 Astra | openai/gpt-6-astra | $10 / $50 | $3 / $15 | −70% |
| GPT 6 Luna | openai/gpt-6-luna | $0.10 / $0.50 | $0.03 / $0.15 | −70% |
| GPT 5.5 | openai/gpt-5.5 | $5 / $30 | $1.50 / $9 | −70% |
| Gemini 3.8 Flash | google/gemini-3.8-flash | $0.75 / $3.75 | $0.225 / $1.125 | −70% |
| Gemini 3.1 Pro Preview | google/gemini-3.1-pro-preview | $2 / $12 | $0.60 / $3.60 | −70% |
| Gemini 2.5 Pro | google/gemini-2.5-pro | $1.25 / $10 | $0.375 / $3 | −70% |
| Gemini 2.5 Flash | google/gemini-2.5-flash | $0.30 / $2.50 | $0.09 / $0.75 | −70% |
| Grok 4.7 | x-ai/grok-4.7 | $2 / $6 | $0.60 / $1.80 | −70% |
| Grok 4.3 | x-ai/grok-4.3 | $1.25 / $2.50 | $0.375 / $0.75 | −70% |
For product-card series, ad creatives and voice-over. Nano Banana on OpenRouter and on API429: the price of one image and of a 10,000-image batch.
| Model and setting | OpenRouter | API429 | 10,000 images on API429 | Difference |
|---|---|---|---|---|
Nano Banana 2 · 1Kgemini-3.1-flash-image | $0.0672 | $0.0201 | $201instead of $672 | −70% |
Nano Banana 2 · 2Kgemini-3.1-flash-image | $0.101 | $0.0303 | $303instead of $1 008 | −69% |
Nano Banana 2 · 4Kgemini-3.1-flash-image | $0.151 | $0.0453 | $453instead of $1 512 | −70% |
Nano Banana 2 Lite · 1Kgemini-3.1-flash-lite-image | $0.0336 | $0.0101 | $101instead of $336 | −70% |
Nano Banana Pro · 1Kgemini-3-pro-image | $0.134 | $0.0402 | $402instead of $1 344 | −70% |
Nano Banana Pro · 2Kgemini-3-pro-image | $0.134 | $0.0402 | $402instead of $1 344 | −70% |
Nano Banana Pro · 4Kgemini-3-pro-image | $0.24 | $0.072 | $720instead of $2 400 | −70% |
OpenRouter's price is its image output token rate (openrouter.ai/api/v1/models, checked October 2, 2026) times Google's documented tokens per image. Input text and references are billed separately.
Per the openrouter.ai/api/v1/models list on October 6, 2026: it has no video generation models, and no FLUX.2 Pro, Grok Imagine or ElevenLabs.
Batches of thousands of images: a job queue, asynchronous 202 responses instead of timeouts, retrying only the failed jobs, and a budget before you start. See the price guide for every Nano Banana setting.
The :free and :online variants are not supported. Remove OpenRouter routing parameters (provider, models, route, transforms) from the request: the gateway picks the route itself.
curl https://gateway.api429.com/v1/chat/completions \
-H "Authorization: Bearer $API429_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "anthropic/claude-fable-5.1",
"messages": [{"role": "user", "content": "Hello!"}]
}'Top up the balance in the dashboard.
For companies, with closing documents.
If you prefer to pay in crypto.
One balance for every model: text, images, video and voice.
Yes. Requests go to the makers' original models, with no lighter or renamed substitutes. The model ID in the response matches the catalog ID.
Usually not. If you use the OpenAI SDK or Chat Completions, change the base URL and the key. OpenRouter-style model IDs are accepted for models in the API429 catalog.
No. Up to 70% applies to the Claude, GPT, Gemini and Grok models in the table above. Open-weight model prices vary; check them in the catalog before moving a workload.
Call GET https://gateway.api429.com/v1/models with your key. A model listed in the catalog is not necessarily enabled for every key.
No. The table compares model rates only. OpenRouter also charges 5.5% on credit purchases on Standard and 8% on Business (openrouter.ai/pricing, October 6, 2026).
Yes, with the same key and balance: Nano Banana, GPT Image, FLUX.2 Pro, Grok Imagine, Veo 3.1 video and voice-over. See the model catalog for the list and prices.
Start with a typical request from your app, check the answer and usage, then move the rest.
OpenRouter is a trademark of its owner. API429 is not affiliated with OpenRouter; the comparison uses published rates.