Models and pricing
Available models, context and price. Input tokens only; output is free.
The Hartley models are our System One decision models, and Hartley 1 is the first. Every Hartley model keeps the same contract: a state in, a calibrated decision for every question out, nothing written.
| Model | Context | Price |
|---|---|---|
Hartley 1 (hartley-1) | 16K tokens | $0.04 per 1M input tokens, output $0 |
Hartley 1 Flash (hartley-1-flash) | 32K tokens | $0.02 per 1M input tokens, output $0 |
hartley-1 is the default when you leave model out of a request. Both models take the same request and return the same answers, so switching is one field.
Which model
| Hartley 1 | Hartley 1 Flash | |
|---|---|---|
| Best for | Hard judgments: agent traces, policy checks, coding-agent verification | High-volume, simpler decisions: routing, tagging, filtering |
| Accuracy | Highest | Lower on hard judgments |
| Context | 16K tokens | 32K tokens |
| Price | $0.04 per 1M input tokens | $0.02 per 1M input tokens |
Start with Hartley 1. Move a question to Flash when it answers it as well on your own examples: the Playground runs either model.
How you're billed
- Input tokens only. Output is free: Hartley doesn't write, so there is no output to bill.
- The state is read once per call. Asking more questions about the same state adds only the questions' own words.
- Failed requests are never billed. Calls are paid from your organization's prepaid credits (Billing and credits), and budgets cap what each key or the whole organization can spend in a month.
Per call$0.0001
Per day$9
Per month$258
GET /v1/models
Lists the models your key can use, with their current prices:
curl https://models.trodo.ai/v1/models -H "Authorization: Bearer $TRODO_API_KEY"{
"data": [
{
"id": "hartley-1",
"object": "model",
"family": "hartley",
"status": "active",
"aliases": [],
"context_tokens": 16384,
"pricing": {"usd_per_million_input_tokens": 0.04, "output": 0}
},
{
"id": "hartley-1-flash",
"object": "model",
"family": "hartley",
"status": "active",
"aliases": ["hartley-flash-latest"],
"context_tokens": 32768,
"pricing": {"usd_per_million_input_tokens": 0.02, "output": 0}
}
]
}How Hartley 1 compares with Jev, CLM-8B and the frontier models, its benchmarks, and its speed: trodo.ai/models/latest.