Skip to main content
Infery gives you access to ~150 models across 10 providers through a single API. Everything is OpenAI-compatible at the wire level, meaning you don’t have to care which provider is behind the slug — just use it.

Live catalog

See the live, browsable catalog at infery.ai/models — filter by modality or provider, see capabilities and context lengths. Or fetch programmatically (auth required):
curl https://api.infery.ai/v1/models \
  -H "Authorization: Bearer $INFERY_API_KEY"
Or without auth (rate-limited to 30 rpm):
curl https://api.infery.ai/public/models

Providers

ProviderModalities
OpenAIChat, Vision, Image (gpt-image-2, DALL·E), TTS, STT (Whisper), Video (Sora), Embeddings
AnthropicChat, Vision, PDF
GoogleChat (Gemini), Vision, Image (Gemini Image / Nano Banana 2, Imagen), Video (Veo), TTS, STT
xAIChat (Grok), Vision, Image, Video
DeepSeekChat (R1, V3)
AlibabaChat (Qwen), Vision, Image (Wan / Qwen Image), Video, TTS, STT
SunoMusic
Self-hostedFLUX, Stable Diffusion XL, Whisper, open embedding models

Common chat slugs

  • OpenAI: gpt-5-4 (flagship), gpt-4o, gpt-4o-mini, o1, o3
  • Anthropic: claude-opus-4-7 (flagship), claude-opus-4-6, claude-sonnet-4-6, claude-haiku-4-5
  • Google: gemini-2-5-pro, gemini-2-5-flash, gemini-2-0-flash
  • xAI: grok-3, grok-3-mini
  • DeepSeek: deepseek-v3, deepseek-r1
  • Alibaba: qwen3-max, qwen3-5-plus, qwen-plus, qwen-flash
Slugs are stable — we don’t rename them. New versions get new slugs (e.g. gpt-4o-2025-04-15 in addition to gpt-4o).

Versioning

Slugs without a date suffix are “latest of that line” (pinned to the currently-best version, updated over time). Slugs with date suffix are frozen — use them for reproducibility when a change in upstream model behaviour would break your app.

Capability filtering

Not every model supports every feature. Some models are chat-only, others vision-enabled, others PDF-native. At app startup, call GET /v1/models/{slug} to get supportsVision, supportsPdf, etc. and branch accordingly. See Capabilities matrix.