# Models & Pricing

Source: https://docs.sliplane.io/inference/models

Models available through Sliplane Inference, their endpoints, and prices.

> **info**
>
> **Private beta:** Email [support@sliplane.io](mailto\:support@sliplane.io) to request Inference access for your team.

Token prices are in EUR per million tokens, excluding VAT. Input prices apply to uncached tokens. Audio transcription is priced per minute.

Every model is available on the Global endpoint. Models marked **EU** are also available on the EU endpoint, with [zero data retention](/inference/getting-started#zero-data-retention). Models marked **Confidential** work with [confidential keys](/inference/keys#confidential-keys).

> **info**
>
> Need a model that isn't listed? Email [support@sliplane.io](mailto\:support@sliplane.io) and we'll check whether we can add it.

## Chat models

Use chat models with `/v1/chat/completions` and `/v1/responses`. All chat models support streaming. For function calls with `gpt-6.1-terra`, use `/v1/responses`.

| Model                  | EU  | Confidential | Input | Cached input | Output |
| ---------------------- | --- | ------------ | ----- | ------------ | ------ |
| `claude-fable-5-1`     |     |              | 10.00 | 0.25         | 50.00  |
| `claude-haiku-5-5`     |     |              | 0.10  | 0.01         | 0.50   |
| `claude-opus-5-5`      |     |              | 4.00  | 0.20         | 20.00  |
| `claude-sonnet-5-5`    |     |              | 2.00  | 0.10         | 10.00  |
| `deepseek-v4-pro-0813` | Yes |              | 1.80  | 0.50         | 3.60   |
| `deepseek-v4.1-flash`  | Yes | Yes          | 0.50  | 0.10         | 1.30   |
| `gemini-3.8-flash`     |     |              | 1.50  | 0.15         | 7.50   |
| `gemini-4-argon`       |     |              | 4.00  | 0.20         | 20.00  |
| `gemma4-26b`           | Yes | Yes          | 0.10  | 0.10         | 0.50   |
| `gemma4-31b`           | Yes |              | 0.40  | 0.40         | 2.00   |
| `glm-5.3`              | Yes | Yes          | 1.50  | 0.35         | 4.00   |
| `glm-5.3-flash`        | Yes |              | 0.25  | 0.05         | 0.50   |
| `gpt-6-astra`          |     |              | 10.00 | 1.00         | 50.00  |
| `gpt-6-luna`           |     |              | 0.10  | 0.01         | 0.50   |
| `gpt-6.1-sol`          |     |              | 2.00  | 0.10         | 10.00  |
| `gpt-6.1-terra`        |     |              | 1.00  | 0.05         | 5.00   |
| `gpt-oss-120b`         | Yes | Yes          | 1.00  | 1.00         | 4.00   |
| `kimi-k3`              | Yes |              | 3.00  | 0.90         | 15.00  |
| `minimax-m3`           | Yes |              | 0.40  | 0.10         | 2.00   |
| `ministral-14b-2512`   | Yes |              | 0.22  | 0.02         | 0.22   |
| `ministral-3b-2512`    | Yes |              | 0.11  | 0.01         | 0.11   |
| `ministral-8b-2512`    | Yes |              | 0.17  | 0.02         | 0.17   |
| `mistral-large-4`      | Yes |              | 1.50  | 0.15         | 4.60   |
| `mistral-medium-3-5`   | Yes |              | 1.65  | 0.17         | 8.25   |
| `mistral-small-4-119b` | Yes | Yes          | 0.50  | 0.50         | 2.00   |
| `qwen3.8-27b`          | Yes | Yes          | 0.50  | 0.12         | 2.50   |

## Transcription models

Use transcription models with `/v1/audio/transcriptions`. Send the audio file as a multipart upload. Streaming is not supported.

| Model              | EU  | Confidential | EUR per audio minute |
| ------------------ | --- | ------------ | -------------------: |
| `whisper-large-v3` | Yes |              |                 0.01 |

The price excludes VAT and is prorated by the reported audio duration. Only delivered transcriptions are charged. See the [request example](/inference/getting-started#transcribe-audio) and [upload limit](/inference/limits#request-limits).

## Embedding models

Use embedding models with `/v1/embeddings`.

| Model                | EU  | Confidential | Input | Cached input |
| -------------------- | --- | ------------ | ----- | ------------ |
| `qwen3-embedding-8b` | Yes |              | 0.03  | 0.03         |

## Decision models

Use decision models with `/v1/systemone`. Only input tokens are billed.

| Model        | EU  | Confidential | Input | Cached input |
| ------------ | --- | ------------ | ----- | ------------ |
| `clef`       |     |              | 0.30  | 0.30         |
| `clef-flash` | Yes |              | 0.15  | 0.15         |

## Additional pricing

* **Long prompts:** for `gpt-6-astra`, `gpt-6.1-sol`, and `gpt-6.1-terra`, prompts above 272,000 tokens cost 2× the input and cached input rates and 1.5× the output rate for the entire request. For `gpt-6-luna`, prompts above 272,000 tokens cost 5× the input, cached input, and output rates for the entire request. For `claude-haiku-5-5`, prompts above 100,000 tokens cost 0.50 input, 0.05 cached input, and 2.50 output for the entire request.
* **Cache writes:** GPT cache writes cost 1.25× the input rate. Claude cache writes cost 1.25× the input rate for five-minute storage and 2× for one-hour storage.