> ## Documentation Index
> Fetch the complete documentation index at: https://docs.corvex.cloud/llms.txt
> Use this file to discover all available pages before exploring further.

# Pricing

> Per-token rates, prepaid billing, and account credits for Corvex Token Factory.

Corvex Token Factory publishes separate rates for input tokens, cached input
tokens, and output tokens. Rates are in US dollars per million tokens.

> [!NOTE]
>
> Inference usage is deducted from your available credits at the published rates
> below. Check your balance and add credits with a card in the
> [Token Factory dashboard](https://tokenfactory.corvex.cloud/app/billing).

## Per-million-token rates

| Model | Input | Cached input | Output | Model details |
| - | - | - | - | - |
| GLM 5.3 | \$1.40 / 1M | \$0.26 / 1M | \$4.40 / 1M | [View model](https://docs.corvex.cloud/models/zai-org/GLM-5.3.md) |
| DeepSeek V4 Flash 0731 | \$0.14 / 1M | \$0.07 / 1M | \$0.28 / 1M | [View model](https://docs.corvex.cloud/models/deepseek-ai/DeepSeek-V4-Flash-0731.md) |

The public model catalog is the programmatic source for current rates. Each
model lists `input_rate_cents_per_mtok`, `cached_input_rate_cents_per_mtok`, and
`output_rate_cents_per_mtok`, and the same prices per token in `pricing`:

```bash theme={null}
curl https://api.tokenfactory.corvex.cloud/v1/models
```

The documentation CI compares the rates on this site with the live catalog
when the catalog is reachable.

## Cached input tokens

When a request starts with a prompt prefix that the model recently processed,
the serving engine reuses that work instead of computing it again. Those input
tokens are cached input tokens, billed at the cached input rate. Caching is
automatic: there is no request parameter, and you do not need `cache_control`.
Requests that share a long, unchanged prefix, such as a system prompt or earlier
conversation turns, get the most cached input tokens.

Cached input pricing applies to requests that start at or after
2026-10-27 12:00 UTC. Earlier requests were billed at the input rate for all input
tokens.

Every response reports the cached share of its input tokens:

| API | Input tokens | Cached input tokens |
| - | - | - |
| Chat Completions, Completions | `usage.prompt_tokens` | `usage.prompt_tokens_details.cached_tokens` |
| Responses | `usage.input_tokens` | `usage.input_tokens_details.cached_tokens` |
| Messages | `usage.input_tokens` + `usage.cache_read_input_tokens` | `usage.cache_read_input_tokens` |

In the Messages API, `input_tokens` does not include the cached input tokens, as
in Anthropic's API.

## Estimate usage cost

Calculate the usage cost of a request as:

```text theme={null}
(uncached input tokens / 1,000,000 × input rate)
+ (cached input tokens / 1,000,000 × cached input rate)
+ (output tokens / 1,000,000 × output rate)
```

Uncached input tokens are the input tokens minus the cached input tokens.

For example, a request on GLM 5.3 with 1,000 input tokens, 600 of them cached,
and 500 output tokens costs `$0.002916`:

```text theme={null}
(400 / 1,000,000 × $1.40) + (600 / 1,000,000 × $0.26) + (500 / 1,000,000 × $4.40)
= $0.000560 + $0.000156 + $0.002200
= $0.002916
```

The same request with no cached input tokens costs `$0.00360`.

The response `usage` object reports the token counts needed for this
calculation. Account-level totals are available in the [Token Factory
dashboard](https://tokenfactory.corvex.cloud/app/usage) and from `GET
/api/v1/usage`.

## Credits and billing

New accounts receive a **one-time \$6 welcome credit**, valid for one calendar
year from issuance. It does not renew when used up.

Accounts created before September 15, 2026 received a **\$2,500 Alpha transition
credit** on September 28, 2026. It expires on March 28, 2027. Check your
dashboard for your current balance and credit activity.

Inference requires an available credit balance. If your credits run out, add
more through the dashboard to resume requests. See
[insufficient credits](https://docs.corvex.cloud/reference/errors.md#insufficient-credits-402) for the API
error and retry guidance.

For dedicated single-tenant GPU capacity, contact Corvex about [Reserved
Endpoints](https://docs.corvex.cloud/reference/faq.md#how-do-i-contact-support).

[Open Corvex Token Factory](https://tokenfactory.corvex.cloud/app)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.