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

# Models

> All available models on the EURI API — text generation, embeddings, image generation, speech-to-text, and text-to-speech.

## Text Generation Models

Use with [`POST /chat/completions`](/api-reference/chat-completions).

### OpenAI

| Model                 | Premium | Context | Best for                                                    |
| --------------------- | ------- | ------- | ----------------------------------------------------------- |
| `gpt-4.1-nano`        | No      | 128K    | Fast, cheap tasks                                           |
| `gpt-4.1-mini`        | No      | 128K    | Balanced quality/cost                                       |
| `gpt-4.1`             | Yes     | 128K    | High-quality reasoning                                      |
| `gpt-4o`              | No      | 128K    | Multimodal, vision                                          |
| `gpt-4o-mini`         | No      | 128K    | Fast multimodal                                             |
| `gpt-5`               | Yes     | 128K    | Frontier reasoning                                          |
| `gpt-5.1`             | Yes     | 128K    | Enhanced GPT-5                                              |
| `gpt-5.2`             | Yes     | 128K    | Latest GPT-5 series                                         |
| `gpt-5.4`             | Yes     | 1.05M   | Flagship reasoning, coding, and agentic work                |
| `gpt-5.3-instant`     | Yes     | 128K    | Fast GPT-5                                                  |
| `gpt-5.4-nano`        | Yes     | 400K    | Cheapest GPT-5.4-class model for simple high-volume tasks   |
| `gpt-5.4-mini`        | Yes     | 400K    | Strong mini model for lower-latency coding and agentic work |
| `o3`                  | Yes     | 200K    | Advanced reasoning                                          |
| `o4-mini`             | Yes     | 200K    | Fast reasoning                                              |
| `openai/gpt-oss-20b`  | No      | 128K    | Open-source 20B                                             |
| `openai/gpt-oss-120b` | No      | 128K    | Open-source 120B                                            |

### Google Gemini

| Model                                 | Premium | Context | Best for            |
| ------------------------------------- | ------- | ------- | ------------------- |
| `gemini-2.0-flash`                    | No      | 1M      | Fast, large context |
| `gemini-2.5-flash`                    | No      | 1M      | Latest flash model  |
| `gemini-2.5-flash-lite-preview-06-17` | No      | 1M      | Ultra-light         |
| `gemini-2.5-pro`                      | No      | 1M      | High quality        |
| `gemini-2.5-pro-preview-06-05`        | No      | 1M      | Preview             |
| `gemini-3-flash`                      | Yes     | 1M      | Next-gen flash      |
| `gemini-3.1-pro`                      | Yes     | 1M      | Next-gen pro        |
| `gemini-3.1-flash-lite`               | Yes     | 1M      | Next-gen lite       |

### Anthropic Claude

| Model               | Premium | Context | Best for         |
| ------------------- | ------- | ------- | ---------------- |
| `claude-sonnet-4`   | Yes     | 200K    | Balanced quality |
| `claude-opus-4`     | Yes     | 200K    | Highest quality  |
| `claude-sonnet-4-5` | Yes     | 200K    | Enhanced Sonnet  |
| `claude-opus-4-5`   | Yes     | 200K    | Enhanced Opus    |
| `claude-haiku-4-5`  | Yes     | 200K    | Fast, affordable |
| `claude-sonnet-4-6` | Yes     | 200K    | Latest Sonnet    |
| `claude-opus-4-6`   | Yes     | 200K    | Latest Opus      |

### Meta Llama (via Groq)

| Model                            | Premium | Context | Best for         |
| -------------------------------- | ------- | ------- | ---------------- |
| `llama-4-scout-17b-16e-instruct` | No      | 128K    | Fast open-source |
| `llama-3.3-70b-versatile`        | No      | 128K    | Versatile tasks  |
| `llama-3.1-8b-instant`           | No      | 128K    | Ultra-fast       |

### Other

| Model                | Premium | Context | Best for          |
| -------------------- | ------- | ------- | ----------------- |
| `qwen/qwen3-32b`     | No      | 128K    | Multilingual      |
| `groq/compound`      | No      | 128K    | Agentic, tool use |
| `groq/compound-mini` | No      | 128K    | Light agentic     |
| `sarvam-m`           | No      | 8K      | Indian languages  |

<Note>**Premium** models require wallet balance (top up at [euron.one/euri](https://euron.one/euri)). Non-premium models use your daily free token quota.</Note>

***

## Embedding Models

Use with [`POST /embeddings`](/api-reference/embeddings).

| Model                                        | Provider | Dimensions | Max Input | Premium |
| -------------------------------------------- | -------- | ---------- | --------- | ------- |
| `text-embedding-3-small`                     | OpenAI   | 1,536      | 8,191     | No      |
| `togethercomputer/m2-bert-80M-32k-retrieval` | Together | 768        | 32,768    | No      |
| `gemini-embedding-001`                       | Google   | 128–3,072  | 2,048     | No      |
| `gemini-embedding-2-preview`                 | Google   | 128–3,072  | 8,192     | Yes     |

<Info>
  **`gemini-embedding-2-preview`** is Google's first natively multimodal embedding model, mapping text, images, video, audio, and PDFs into a unified embedding space. Through the Euri API, text embedding is supported. Trained with Matryoshka Representation Learning — set `dimensions` to 768, 1536, or 3072 to balance cost vs. quality.
</Info>

***

## Image Generation Models

Use with [`POST /images/generations`](/api-reference/image-generation).

| Model                        | Provider | Premium |
| ---------------------------- | -------- | ------- |
| `gemini-3-pro-image-preview` | Google   | No      |

***

## Speech-to-Text Models

Use with [`POST /audio/transcriptions`](/api-reference/speech-to-text).

| Model                    | Provider  | Premium |
| ------------------------ | --------- | ------- |
| `whisper-large-v3`       | OpenAI    | Yes     |
| `whisper-large-v3-turbo` | OpenAI    | Yes     |
| `sarvam-stt`             | Sarvam AI | Yes     |

***

## Text-to-Speech Models

Use with [`POST /audio/speech`](/api-reference/text-to-speech).

| Model                             | Provider    | Premium |
| --------------------------------- | ----------- | ------- |
| `canopylabs/orpheus-v1-english`   | Canopy Labs | No      |
| `canopylabs/orpheus-arabic-saudi` | Canopy Labs | No      |
| `playai-tts`                      | PlayAI      | No      |
| `playai-tts-arabic`               | PlayAI      | No      |
| `sarvam-tts`                      | Sarvam AI   | Yes     |

***

## Listing models via API

```bash theme={null}
curl https://api.euron.one/api/v1/euri/models \
  -H "Authorization: Bearer YOUR_EURI_API_KEY"
```