API2D records 214 monthly visits with -91.3% month-over-month growth and Paid pricing. OfoxAI records 239,718 monthly visits with +12.7% growth and Flexible pricing.
Use top AI models like electricity — register, top up, paste your key
One API key, 100+ LLMs, official pricing with no platform markup
| Dimension / Metric | | |
|---|---|---|
| Official Domain | api2d.com | ofox.ai |
| Pricing Model | Paid | Flexible |
| Monthly Visits | 214 (214) | 239.72K (239,718) |
| MoM Traffic Delta | -2.23K (-91.3%) | +27.06K (+12.7%) |
| Primary Market & Share | 🇹🇼 Taiwan (100%) | 🇨🇳 China (24.5%) |
| Direct Visits / mo | 86 | 81.89K |
| User Rating | 3.8 | 4.4 |
| Supported Platforms | Website | Website |
Use top AI models like electricity — register, top up, paste your key
One base URL and key work with virtually any app, SDK or self-hosted chat client that supports a custom API address.
Route requests to OpenAI models, Claude, Stability diffusion, Azure TTS and vector database services from a single account.
Consumption is metered in points with a published multiplier, so you pay for what you use rather than a flat subscription.
Identical request bodies return cached answers for free for 24 hours, and a header can disable caching when freshness matters.
One API key, 100+ LLMs, official pricing with no platform markup
Reach text, image, video, embedding, speech, and transcription models from OpenAI, Anthropic, Google, DeepSeek, Qwen, GLM, Kimi, and more through one key and one endpoint, with new models added shortly after release.
Point your existing OpenAI, Anthropic, or Gemini SDK at ofox.ai's base URL and keep your code. No proprietary client library has to be adopted.
Pay the model provider's official list price with a stated 0% platform fee, and let automatic volume credits of up to about 7% kick in as spend grows. No monthly subscription is required.
Set daily, monthly, or total budgets per team, member, or API key, with requests blocked over the limit. Dashboards break cost, tokens, and cache usage down by model, key, and member.
SpecialistFocuses specifically on native Google Gemini models and prototyping tools rather than multi-provider aggregation.
SpecialistFocuses specifically on native Google Gemini models and prototyping tools rather than multi-provider aggregation.
Offers massive model selection and robust routing, but lacks native multi-modal media generation endpoints.
Offers massive model selection and robust routing, but lacks native multi-modal media generation endpoints.
Free / Open-sourceOne OpenAI-compatible gateway for adaptive LLM routing, guardrails, and spend control
Free / Open-sourceOne OpenAI-compatible gateway for adaptive LLM routing, guardrails, and spend control
Extends text routing with robust media inference endpoints, but has a different pricing structure.
Extends text routing with robust media inference endpoints, but has a different pricing structure.
Provides access to 500+ models under an OpenAI-compatible endpoint, but with less community documentation.
Provides access to 500+ models under an OpenAI-compatible endpoint, but with less community documentation.
Free / Open-sourceProvides self-hostable open-source proxy and Python SDK, but requires manual server setup and API keys.
Combines a unified LLM API gateway with a built-in multi-model chat interface for testing.
Combines a unified LLM API gateway with a built-in multi-model chat interface for testing.
Free / Open-sourceOne API for LLM, image, video, and audio models in production
Free / Open-sourceOne API for LLM, image, video, and audio models in production
Provides similar unified OpenAI-compatible routing for diverse LLMs, with automated failover and discounted pricing.
Aggregates hundreds of foundational models and agent tools, focusing on multi-vendor developer integration.
One OpenAI-compatible key for discounted GPT, Claude and Gemini access
Free / Open-sourceAn open-source routing gateway with built-in guardrails, but requires self-hosting infrastructure.
Free / Open-sourceAn open-source routing gateway with built-in guardrails, but requires self-hosting infrastructure.
