Overview
Models
OneRouter provides access to 200+ models from 50+ providers through a single OpenAI-compatible endpoint. Browse by category or search for specific models.
Model Categories
Large Language Models
State-of-the-art text generation models for chat, reasoning, code generation, and content creation.
| Model | Provider | Context | Best For |
|---|---|---|---|
| GPT-4o | OpenAI | 128K | General purpose, multimodal |
| GPT-4.1 | OpenAI | 1M | Long context, coding |
| Claude Opus 4.8 | Anthropic | 200K | Complex reasoning, analysis |
| Claude Sonnet 4.6 | Anthropic | 200K | Balanced speed & quality |
| Gemini 2.5 Pro | 1M | Long context, research | |
| DeepSeek V3 | DeepSeek | 128K | Cost-efficient, coding |
| Llama 4 | Meta | 128K | Open-source alternative |
| Mistral Large 3 | Mistral | 128K | European AI, multilingual |
Vision & Multimodal
Models that understand images, screenshots, and documents alongside text.
| Model | Provider | Input |
|---|---|---|
| GPT-4o Vision | OpenAI | Text + Image |
| Claude 4 Vision | Anthropic | Text + Image + PDF |
| Gemini 2.5 Vision | Text + Image + Video | |
| Qwen-VL | Alibaba | Text + Image |
Embeddings & Reranking
Vector embeddings for semantic search, RAG, clustering, and similarity detection.
| Model | Provider | Dimensions |
|---|---|---|
| text-embedding-3-large | OpenAI | 3072 / 256-1024 |
| text-embedding-3-small | OpenAI | 1536 / 512 |
| Cohere Embed v3 | Cohere | 1024 |
| Voyage AI | Voyage | 1024 / 2048 |
| BGE-M3 | BAAI | 1024 |
Speech & Audio
Text-to-speech and speech-to-text models for voice applications.
| Model | Provider | Type |
|---|---|---|
| GPT-4o TTS | OpenAI | Text-to-Speech |
| Whisper | OpenAI | Speech-to-Text |
| ElevenLabs | ElevenLabs | Text-to-Speech |
Switching Models
To use any model, simply pass its ID in the model field of your API request:
# Switch from GPT-4o to Claude Opus — same code, one param change
response = client.chat.completions.create(
model="claude-opus-4-8", # ← Just change this
messages=[{"role": "user", "content": "Write a haiku about coding."}],
)Browse the complete model catalog with live pricing at api.onerouter.app/pricing.
Reasoning Effort Control
For reasoning models (GPT-5, Claude Opus, DeepSeek R1, Qwen3), you can control the depth of reasoning by appending a suffix to the model name:
| Suffix | Effect | Example |
|---|---|---|
-high | Maximum reasoning depth — best quality, slower | gpt-5-high |
-medium | Balanced reasoning (default for most models) | gpt-5-medium |
-low | Fast, minimal reasoning — best for simple tasks | gpt-5-low |
-thinking | Returns reasoning tokens in the response | claude-opus-4-8-thinking |
If no suffix is provided, the model's default reasoning level is used. This lets you trade off speed and cost against answer quality per request, without changing your integration code.
Cross-Format Conversion
OneRouter automatically converts between different API formats behind the scenes. You always send requests in the OpenAI Chat Completions format — OneRouter handles translation to the upstream provider's native format:
- OpenAI → Claude Messages API — System prompts, multi-turn conversation, and tool definitions are automatically remapped
- OpenAI → Gemini API — Content parts, safety settings, and generation config are translated transparently
- Claude → OpenAI format — Claude's native responses are normalized to the standard Chat Completions response structure
This means you can use the OpenAI Python/Node SDK, LangChain, or any OpenAI-compatible library to call any model — regardless of its native API format.