FAQ
Frequently asked questions
Quick answers to the questions we hear most often about Elevence AI.
Applies to web, iOS and Android
General
What is Elevence AI?
Elevence AI is a unified multi-model AI studio. It brings 60+ models from OpenAI, Anthropic, Google, Meta, Mistral, ElevenLabs, FAL, Runway, Luma, Perplexity, xAI and others behind one OpenAI-compatible API and one consumer studio that runs on iOS, Android and Web from a single Expo / React Native codebase. Built by NextGen AI Dev Private Limited.
How is Elevence AI priced?
Pay-as-you-go credits. Each request deducts credits proportional to the underlying provider rate plus a transparent platform margin. Free starter credits on sign-up. No subscription required. Pro and Team plans offer subscription credits with rollover; Enterprise plans offer custom pricing and volume discounts.
Is Elevence AI an OpenAI-compatible API?
Yes. The /v1/chat/completions endpoint accepts the exact same request shape as OpenAI's official API. Point any OpenAI SDK at https://api.elevence.ai/v1 with your Elevence API key and the code keeps working. Change the model field to swap between GPT-5, Claude, Gemini, Llama, Mistral and the other 55+ models behind one key.
Does Elevence AI use my data for training?
No. Prompts and outputs are private to your account and are not sent to providers for training. Standard data-retention controls apply across providers. You can delete any conversation or your entire history at any time.
Which AI models does Elevence AI support?
Text: GPT-5, Claude Opus/Sonnet/Haiku 4.x, Gemini 3 Pro/Flash, Llama 4, Mistral Large, Cohere Command, Perplexity Sonar, plus open-source variants via Groq, Together, HuggingFace. Image: FLUX, Stable Diffusion, DALL·E 3, Ideogram, Leonardo, Vertex. Video: Luma Dream Machine, Runway Gen-3, Google Veo, Kling. Audio: ElevenLabs (TTS, STT, Music), OpenAI TTS, Whisper, Suno. 60+ models in total, refreshed as new releases ship.
Can I run Elevence AI on iOS or Android?
Yes. Elevence AI ships native iOS and Android apps from the same Expo / React Native 0.81 codebase as the web studio. Conversations, knowledge bases, agents and credits sync across devices.
Is there an enterprise plan with SSO and audit logs?
Yes. The Enterprise plan includes SSO via SAML/OIDC, audit logging, custom model routing, signed Data Processing Agreement, volume pricing, and dedicated support. See /enterprise for details.
How does Elevence AI compare to ChatGPT, Claude or Perplexity?
Elevence AI is broader — 60+ models instead of one vendor, image/video/audio generation alongside chat, pay-as-you-go vs subscription, and an OpenAI-compatible API for developers. See /alternatives for head-to-head comparisons against each major AI app.
Model-specific questions
Each AI model has its own dedicated page with full FAQ. The most common questions across all models:
How is GPT-5 different on Elevence AI vs OpenAI direct?
Yes — Elevence AI calls OpenAI's official API, so you get exactly the same GPT-5 with the same training data, capabilities and reasoning effort tiers. The only difference is the wrapper (one key for 60+ models) and pay-as-you-go credit billing.
How is Claude Opus & Sonnet different on Elevence AI vs Anthropic direct?
Default to Sonnet 4.6 for production — best quality-per-dollar ratio. Upgrade to Opus 4.7 for the hardest reasoning tasks where quality is worth 5x the cost. Drop to Haiku 4.5 for high-throughput simple tasks (classification, extraction).
How is Gemini 3 Pro & Flash different on Elevence AI vs Google direct?
At $0.10/$0.40 per million in/out tokens, processing a million-token input + a 10K-token output costs roughly $0.10 + $0.004 = ~$0.10 per request. For batch pipelines processing thousands of long documents, Flash is dramatically cheaper than GPT-5 at $1.25/$10.
Use-case questions
Is Elevence AI compatible with the official OpenAI SDK?
Yes. Both the Python `openai` and Node `@openai/sdk` clients work unchanged — just set `base_url` (or `baseURL`) to `https://api.elevence.ai/v1` and pass your Elevence API key. Streaming, tool calls, vision and multi-turn all behave identically.
Can I keep visual style consistent across image and video?
Yes. Generate a hero still with FLUX or DALL·E, then pass it as the first frame to Runway or Luma — the video starts from that exact image and preserves the style. Generate multiple variant clips with the same starting frame for shot variation.
Can the AI learn our brand voice from existing content?
Yes. Upload 5-20 examples of brand-voice writing to a knowledge base, then reference it in any chat. The model retrieves and applies the voice consistently. For visual brand identity, use a reusable style prompt or a custom LoRA for image generation.
Alternatives & comparisons
How does Elevence AI compare to ChatGPT?
How Elevence AI compares to OpenAI's ChatGPT for creators, developers and teams that need more than one model.
How does Elevence AI compare to Claude.ai?
How Elevence AI compares to Anthropic's Claude.ai for long-context work, coding and multi-model workflows.
How does Elevence AI compare to Perplexity?
How Elevence AI compares to Perplexity for search-first AI workflows and research.