# x402 AI API — zeroreader.com
# Pay-per-request AI inference via x402 micropayments (USDC on Base)

> 29 AI models available. No API keys needed.

## Models

### Llm
- [IBM Granite 4.0 Micro](https://api.zeroreader.com/v1/ai/granite-micro): Ultra-cheap, fast responses. Best for simple tasks. ($0.001/request)
- [Llama 3.2 1B](https://api.zeroreader.com/v1/ai/llama-1b): Smallest Llama. Fastest, cheapest. ($0.001/request)
- [Llama 3.2 3B](https://api.zeroreader.com/v1/ai/llama-3b): Good balance of speed and quality for simple tasks. ($0.002/request)
- [Llama 3.1 8B (Fast)](https://api.zeroreader.com/v1/ai/llama-8b): Best value. Fast inference, solid quality. ($0.002/request)
- [Mistral 7B](https://api.zeroreader.com/v1/ai/mistral-7b): European open-source model. Good at structured output. ($0.002/request)
- [Qwen3 30B (MoE, 3B active)](https://api.zeroreader.com/v1/ai/qwen3-30b): 30B MoE model with only 3B active params. Fast + smart. ($0.003/request)
- [GPT-OSS 20B (OpenAI)](https://api.zeroreader.com/v1/ai/gpt-oss-20b): OpenAI's open-source 20B model. ($0.002/request)
- [GLM 4.7 Flash](https://api.zeroreader.com/v1/ai/glm-flash): Zhipu AI model. Strong multilingual. ($0.002/request)
- [Llama 3.2 11B Vision](https://api.zeroreader.com/v1/ai/llama-vision): Vision + text model. Can understand images. ($0.005/request)
- [Llama 4 Scout 17B](https://api.zeroreader.com/v1/ai/llama-4-scout): Latest Llama 4 architecture. 17B with 16 experts. ($0.005/request)
- [Mistral Small 3.1 24B](https://api.zeroreader.com/v1/ai/mistral-24b): Strong mid-size European model. ($0.004/request)
- [Gemma 3 12B (Google)](https://api.zeroreader.com/v1/ai/gemma-12b): Google's open model. Good at reasoning. ($0.004/request)
- [SEA-LION 27B](https://api.zeroreader.com/v1/ai/sea-lion-27b): Southeast Asian languages specialist. ($0.004/request)
- [Llama 3.3 70B (Fast)](https://api.zeroreader.com/v1/ai/llama-70b): Flagship open-source LLM. Best quality on CF Workers AI. ($0.008/request)
- [Qwen 2.5 Coder 32B](https://api.zeroreader.com/v1/ai/qwen-coder-32b): Code generation specialist. Excellent for programming tasks. ($0.006/request)
- [QwQ 32B (Reasoning)](https://api.zeroreader.com/v1/ai/qwq-32b): Reasoning-focused model from Qwen. ($0.006/request)
- [GPT-OSS 120B (OpenAI)](https://api.zeroreader.com/v1/ai/gpt-oss-120b): OpenAI's largest open-source model. ($0.010/request)
- [DeepSeek R1 32B (Reasoning)](https://api.zeroreader.com/v1/ai/deepseek-r1): Chain-of-thought reasoning. Best for math/logic/code. ($0.015/request)

### Embedding
- [BGE-M3 (Multilingual)](https://api.zeroreader.com/v1/ai/embed-bge-m3): Best multilingual embedding model. 100+ languages. ($0.001/request)
- [BGE Large EN v1.5](https://api.zeroreader.com/v1/ai/embed-bge-large): High-quality English embeddings. ($0.002/request)
- [BGE Small EN v1.5](https://api.zeroreader.com/v1/ai/embed-bge-small): Fast, lightweight English embeddings. ($0.001/request)
- [Qwen3 Embedding 0.6B](https://api.zeroreader.com/v1/ai/embed-qwen3): Compact embedding model from Qwen. ($0.001/request)

### Image
- [FLUX.1 Schnell](https://api.zeroreader.com/v1/ai/flux-schnell): Fast image generation (~2s). Good for prototyping. ($0.010/request)

### Audio
- [Whisper (Speech-to-Text)](https://api.zeroreader.com/v1/ai/whisper): OpenAI Whisper. Transcription in 90+ languages. ($0.008/request)
- [Whisper Large v3 Turbo](https://api.zeroreader.com/v1/ai/whisper-large): Latest Whisper. Better accuracy, faster. ($0.010/request)
- [MeloTTS (Text-to-Speech)](https://api.zeroreader.com/v1/ai/tts): Text to speech synthesis. ($0.005/request)

### Other
- [M2M100 1.2B (Translation)](https://api.zeroreader.com/v1/ai/translate): 100-language translation model. ($0.002/request)
- [BGE Reranker](https://api.zeroreader.com/v1/ai/reranker): Re-rank search results for better relevance. ($0.001/request)
- [DistilBERT Sentiment](https://api.zeroreader.com/v1/ai/sentiment): Sentiment analysis (positive/negative). ($0.001/request)

## Protocol
- Payment: x402 (USDC on Base, network eip155:8453)
- Discovery: GET /.well-known/x402.json
- Agent Info: GET /.well-known/agent.json
- Model List: GET /v1/models
- OpenAPI Spec: GET /openapi.json