# NezhaGate

> NezhaGate (a.k.a. NezhaAPI) is an OpenAI-compatible AI model gateway: one endpoint and one API key for GPT, Claude, Gemini and image / video generation models. Pay-as-you-go; failed requests are never billed. Base URL: https://nezhagate.com/v1. Last updated 2026-09-11.

## Getting started
- [Integration guide](https://nezhagate.com/en/docs-guide): sign up, create an API key in the console, point base_url at https://nezhagate.com/v1; the official OpenAI SDK, Python, Node.js, Go and curl all work as-is
- [FAQ](https://nezhagate.com/en/faq): 25 answers on billing, compatibility and acceptable use; works with n8n / Dify / LangChain / Open WebUI / LobeChat / Cherry Studio / ChatBox

## Models & live pricing
- [Model market](https://nezhagate.com/en/market): every model on sale (chat / image / video)
- [Pricing](https://nezhagate.com/en/pricing): per-model prices vs the official list price
- [GPT-5.6 Sol](https://nezhagate.com/en/model/gpt-5.6-sol): OpenAI chat / reasoning model; input $2 / output $12 per 1M tokens (official $8/$48, ~75% cheaper)
- [GPT-5.6 Terra](https://nezhagate.com/en/model/gpt-5.6-terra): OpenAI chat / reasoning model; input $1.2 / output $7 per 1M tokens (official $5/$30, ~77% cheaper)
- [GPT-5.6 Luna](https://nezhagate.com/en/model/gpt-5.6-luna): OpenAI chat / reasoning model; input $0.8 / output $4.8 per 1M tokens (official $2.5/$15, ~68% cheaper)
- [GPT-5.5](https://nezhagate.com/en/model/gpt-5.5): OpenAI chat / reasoning model; input $0.7 / output $4.2 per 1M tokens (official $5/$30, ~86% cheaper)
- [GPT-6 Astra](https://nezhagate.com/en/model/gpt-6-astra): OpenAI chat / reasoning model; input $2.8 / output $14 per 1M tokens (official $10/$50, ~72% cheaper)
- [GPT Image 2](https://nezhagate.com/en/model/gpt-image-2): OpenAI image generation model; $0.015 (1K) / $0.025 (2K) / $0.04 (4K) per image
- [GPT Image 2.5 Flare](https://nezhagate.com/en/model/gpt-image-2.5-flare): OpenAI image generation model; $0.015 (1K) / $0.025 (2K) / $0.04 (4K) per image
- [GPT Image 2.5 Sunburst](https://nezhagate.com/en/model/gpt-image-2.5-sunburst): OpenAI image generation model; $0.015 (1K) / $0.025 (2K) / $0.04 (4K) per image
- [Nano Banana 2](https://nezhagate.com/en/model/nano-banana-2): Google image generation model; $0.025 (1K) / $0.0375 (2K) / $0.0625 (4K) per image
- [Nano Banana Pro](https://nezhagate.com/en/model/nano-banana-pro): Google image generation model; $0.04 (1K) / $0.06 (2K) / $0.1 (4K) per image
- [Claude Sonnet 4.6](https://nezhagate.com/en/model/claude-sonnet-4-6): Anthropic chat / reasoning model; input $1.5 / output $7.5 per 1M tokens (official $3/$15, ~50% cheaper)
- [Claude Opus 5](https://nezhagate.com/en/model/claude-opus-5): Anthropic chat / reasoning model; input $4 / output $20 per 1M tokens
- [Claude Fable 5](https://nezhagate.com/en/model/claude-fable-5): Anthropic chat / reasoning model; input $8 / output $40 per 1M tokens
- [Gemini 3.1 Pro](https://nezhagate.com/en/model/gemini-3.1-pro): Google chat / reasoning model; input $0.5 / output $3 per 1M tokens (official $2/$12, ~75% cheaper)
- [Gemini 3.8 Flash](https://nezhagate.com/en/model/gemini-3.8-flash): Google chat / reasoning model; input $0.6 / output $3.6 per 1M tokens (official $0.75/$3.75, ~7% cheaper)
- [Gemini 3.7 Flash](https://nezhagate.com/en/model/gemini-3.7-flash): Google chat / reasoning model; input $0.6 / output $3.6 per 1M tokens (official $0.75/$3.75, ~7% cheaper)
- [Gemini 3.6 Flash](https://nezhagate.com/en/model/gemini-3.6-flash): Google chat / reasoning model; input $0.6 / output $3.6 per 1M tokens
- [Gemini 3.6 Flash High](https://nezhagate.com/en/model/gemini-3.6-flash-high): Google chat / reasoning model; input $0.6 / output $3.6 per 1M tokens
- [Gemini 3.6 Flash Low](https://nezhagate.com/en/model/gemini-3.6-flash-low): Google chat / reasoning model; input $0.6 / output $3.6 per 1M tokens
- [Gemini 3.6 Flash Tiered](https://nezhagate.com/en/model/gemini-3.6-flash-tiered): Google chat / reasoning model; input $0.6 / output $3.6 per 1M tokens
- [Gemini 3 Flash](https://nezhagate.com/en/model/gemini-3-flash-preview): Google chat / reasoning model; input $0.3 / output $1.2 per 1M tokens (official $0.5/$3, ~57% cheaper)
- [Gemini 2.5 Flash](https://nezhagate.com/en/model/gemini-2.5-flash): Google chat / reasoning model; input $0.3 / output $1.2 per 1M tokens (official $0.3/$2.5, ~46% cheaper)
- [Veo 3.1](https://nezhagate.com/en/model/veo-3.1): Google video generation model; from $0.075 per clip (priced by tier / resolution); no charge on failure, the hold is refunded in full
- [Seedance 2.5](https://nezhagate.com/en/model/seedance-2.5): ByteDance video generation model; $0.158 per second (4-30s, billed per second); no charge on failure, the hold is refunded in full
- [Seedance 2.0](https://nezhagate.com/en/model/seedance-2.0): ByteDance video generation model; $0.15 per second (5-15s, billed per second); no charge on failure, the hold is refunded in full
- [Seedance 2.0 Fast](https://nezhagate.com/en/model/seedance-2.0-fast): ByteDance video generation model; $0.1 per second (5-15s, billed per second); no charge on failure, the hold is refunded in full
- [Seedance 2.0 Mini](https://nezhagate.com/en/model/seedance-2.0-mini): ByteDance video generation model; $0.066 per second (5-15s, billed per second); no charge on failure, the hold is refunded in full
- [Wan 3.0](https://nezhagate.com/en/model/wan3.0-video): Alibaba video generation model; $0.12 per second (2-30s, billed per second); no charge on failure, the hold is refunded in full
- [Wan 3.0 Prime](https://nezhagate.com/en/model/wan3.0-video-prime): Alibaba video generation model; $0.16 per second (2-30s, billed per second); no charge on failure, the hold is refunded in full
- [MiniMax H3](https://nezhagate.com/en/model/minimax-h3): MiniMax video generation model; $0.036 per second (5-15s, billed per second); no charge on failure, the hold is refunded in full
- [Grok Imagine Video 1.5](https://nezhagate.com/en/model/grok-imagine-video-1.5): xAI video generation model; from $0.3 per clip (priced by tier / resolution); no charge on failure, the hold is refunded in full

## Docs
- [Gemini 3.6 Flash examples](https://nezhagate.com/en/docs/gemini-3.6-flash): copy-paste curl / Python / Node.js
- [GPT-5.5 examples](https://nezhagate.com/en/docs/gpt-5.5): copy-paste curl / Python / Node.js
- [Veo 3.1 video](https://nezhagate.com/en/docs/veo-3.1): text/image-to-video, 720p-4K, per-clip billing with a full refund on failure

## Optional
- [Full reference llms-full.txt](https://nezhagate.com/llms-full.txt): code samples + full price table + FAQ
- [中文版 Chinese version](https://nezhagate.com/llms.txt)
- [Versión en español](https://nezhagate.com/es/llms.txt)
- [한국어판 Korean version](https://nezhagate.com/ko/llms.txt)
- [日本語版 Japanese version](https://nezhagate.com/ja/llms.txt)
- [Terms](https://nezhagate.com/en/terms)
- [Privacy](https://nezhagate.com/en/privacy)