Models
Proxyify gives access to 300+ models across every major provider. All models are referenced by a provider/model-name slug.
Model naming
Every model is identified by a slug in the format provider/model-name:
examples
openai/gpt-4o
openai/gpt-4o-mini
anthropic/claude-3-5-sonnet
anthropic/claude-3-5-haiku
google/gemini-2.0-flash
google/gemini-2.5-pro
meta-llama/llama-4-scout
black-forest-labs/flux-1.1-pro
kling/kling-video-v3-pro
openai/whisper-1
openai/gpt-4o-mini-tts
Pass the slug directly in the model field of your request. The provider prefix is required — gpt-4o alone will not resolve.
Modalities
| Modality | Description | Example models |
|---|---|---|
text | Chat completions, instruction following, reasoning | openai/gpt-4o, anthropic/claude-3-5-sonnet |
image | Image generation from text prompts | black-forest-labs/flux-1.1-pro, recraft-ai/recraft-v3 |
video | Video generation (async, polling) | kling/kling-video-v3-pro, google/veo-3 |
stt | Speech-to-text transcription | openai/whisper-1, google/chirp-3 |
tts | Text-to-speech audio generation | openai/gpt-4o-mini-tts, hexgrad/kokoro-82m |
When creating an API key, you can restrict it to specific modalities using allowed_categories. For example, a key with allowed_categories: ["text"] cannot be used for image or video generation.
Browse models
The full list of supported models is available in the dashboard. You can filter by modality, search by name, and click any slug to copy it to clipboard.