For developers
What Smarty Pants inference costs
Every model has a listed price per million tokens, the same unit most providers use. Here are the prices on November 19, 2026, and a calculator for your own traffic.
Short answer: On November 19, 2026, Smarty Pants listed released chat models from $0.027 per million input tokens, embeddings at $0.01 per million tokens, speech to text at $0.18 an hour of audio and images at $0.003 each. You pay only for requests it serves; the live price list is GET /v1/models.
Listed prices
| Model | What it does | Input | Output | Status |
|---|---|---|---|---|
| Qwen3 Embedding 0.6B | Embeddings | $0.01 / M tokens | — | Released |
| Qwen3.5 9B | Chat | $0.10 / M tokens | $0.40 / M tokens | Released |
| gpt-oss-20b | Chat | $0.027 / M tokens | $0.135 / M tokens | Released |
| Nanbeige4.3 33B | Chat | $0.30 / M tokens | $2.00 / M tokens | Released |
| Qwen3.6 35B-A3B | Chat | $0.09 / M tokens | $0.80 / M tokens | Released |
| Qwen3.8 27B | Chat | $0.20 / M tokens | $1.20 / M tokens | Released |
| Qwen3.5 4B | Chat | $0.10 / M tokens | $0.25 / M tokens | In testing, opens soon |
| Whisper large-v3 turbo | Speech to text | $0.18 an hour of audio | In testing, opens soon | |
| FLUX.1 schnell | Image making | $0.003 a picture | In testing, opens soon | |
Source: GET /v1/models and smartypants.network/api, checked November 19, 2026. "M tokens" means one million tokens; for English text, a token is about three-quarters of a word (other languages and code use more tokens per word).
Estimate your monthly cost
Enter your own traffic. The calculator multiplies it by the listed price; it runs in your browser and sends nothing anywhere.
Reading these prices
Input and output are priced separately because writing tokens takes more work than reading them. For chat, output usually dominates the bill when answers are long; for retrieval and classification, input does. LLM API pricing explained walks through the math, and Smarty Pants vs. centralized providers covers when a GPU cloud is the better buy.
Frequently asked questions
How does Smarty Pants charge?
Per token, at the listed price per million input tokens and per million output tokens for the model you call. Speech to text is priced per hour of audio and image making per picture. You are not charged for a request Smarty Pants can't serve.
Where is the current price?
GET https://smartypants.network/v1/models returns every model's price (per token, in its pricing field). The prices on this page were checked on November 19, 2026 and can change.
Do reasoning tokens cost extra?
Reasoning tokens count as output. Each model's default_reasoning field in /v1/models says whether reasoning is on by default; turning it off keeps output short.
Is there a minimum spend or subscription?
Smarty Pants does not list a subscription for the API. You add credit to a balance and requests are paid from it; a 402 error means the balance is too low.
How accurate is the calculator?
It multiplies your inputs by the listed prices, nothing more. Real token counts depend on the tokenizer and your prompts, so measure a sample of your own requests before you budget.
Try a pinned open model on real Macs.
See every model, its listed price and the code to call it. Smarty Pants works with the OpenAI SDK you already use: change the base URL and the key.
Requests run on Macs owned by other people. Owners agree not to read requests and the app doesn't show them, but a serving Mac decrypts each request in memory to answer it, so that is a rule and a design, not a technical guarantee. Don't send sensitive data on the general network.