Major providers
OpenAI
GPT-6.2 Astra and Sol, GPT-6 Astra, the GPT-5.6 series (Sol, Terra, Luna), GPT-5.5, GPT-5.4 and GPT-5.4 mini, plus GPT Image 2 for image generation.
Anthropic Claude
Claude Fable, Opus, Sonnet and Haiku.
Google Gemini
Gemini 4 Argon, Gemini Pro, Flash, Flash-Lite and the image generation models (Nano Banana).
DeepSeek
DeepSeek V4 Pro and DeepSeek Flash — affordable reasoning models.
Zhipu GLM
glm-5.3, glm-5.3-flash and GLM-5.2 — text only.
Alibaba Qwen
Qwen4 Max, Qwen3.8 Max, Flash and 27B, Qwen3.7 Max and Plus.
Moonshot Kimi
Kimi K3.1 Pro — ID
kimi-k3.1-pro; faster version — kimi-k3.1-pro-highspeed. 1M-token context. The previous model, Kimi K3, is kimi/kimi-k3.MiniMax
MiniMax M3.1 Flash — ID
MiniMax-M3.1-Flash: a low-cost model for agents and coding with a 1M-token context. MiniMax-M3 is also available — ID MiniMax/MiniMax-M3.ByteDance Seedance
Seedance 2.0 and 2.5 — video generation with sound.
Model IDs must be typed exactly as they appear in the catalog, prefix and letter case included (
kimi/kimi-k3, GLM-5.2, glm-5.3). You can copy them from the Pricing page at www.ruapi.ai.Model capabilities
The catalog covers different use cases:- Chat & text: classic LLMs for conversation, code, analysis.
- Multimodal (vision): models that understand images — see vision.
- Tool use / function calling: invoke external tools from model output — see function calling.
- Reasoning: chain-of-thought reasoning models.
- Image generation: Gemini Image and GPT Image — see image generation.
- Embeddings:
text-embedding-3-smallandtext-embedding-3-large— for RAG and document search, via/v1/embeddings— see the request example and LangChain. - Video generation: Seedance 2.0 and 2.5 — see video generation.
Why RuAPI
One key for every provider
No separate accounts at OpenAI, Anthropic and others — one key works with every provider in the list.
Genuine models
Straight from the providers, no substitution or degradation.
Low latency
Optimized request routing for fast response times.