{"x402Version":2,"name":"Prism402","description":"OpenAI-compatible inference router on Arc. Every model behind one endpoint, paid per call in USDC.","homepage":"https://www.prism402.com","mode":"live","payment_terms":"Chat calls quote a ceiling: the full input plus max_tokens at the posted rate. The ceiling settles through the facilitator and the unused part is owed back to the payer. Prepaid keys (POST /v1/keys) are charged the exact token cost.","settlement":{"chain":"arc","network":"eip155:5042","asset":"USDC","asset_address":"0x3600000000000000000000000000000000000000","decimals":6,"facilitator":"https://facilitator.arcusnetwork.co"},"auth":["x402_pay_per_call","prepaid_api_key"],"endpoints":[{"path":"/v1/chat/completions","method":"POST","type":"openai_compatible_chat","pricing_model":"per_token","amount_semantics":"per_request_ceiling"},{"path":"/v1/embeddings","method":"POST","type":"openai_compatible_embeddings","pricing_model":"per_token","amount_semantics":"exact"},{"path":"/v1/scan","method":"POST","type":"token_risk_scan","pricing_model":"fixed","amount":"2000","params":["address"]},{"path":"/v1/keys","method":"POST","type":"prepaid_key_deposit","pricing_model":"deposit","params":["amount_usd"]}],"accepts":[{"scheme":"exact","network":"eip155:5042","amount":"1","asset":"0x3600000000000000000000000000000000000000","payTo":"0x385F647CCF80C78c407C80578Ddb7fe60d747D61","maxTimeoutSeconds":120,"extra":{"name":"USDC","version":"2"}}],"models":[{"id":"kimi-k2.6","provider":"Moonshot AI","type":"chat","modality":"text+vision","context_length":262144,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":1.19,"output_usdc_per_mtok":5,"provider_name":"Moonshot AI","description":"1T-parameter open model, tool calls and vision."},{"id":"kimi-k2.7-code","provider":"Moonshot AI","type":"chat","modality":"text","context_length":262144,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":1.19,"output_usdc_per_mtok":5,"provider_name":"Moonshot AI","description":"Kimi tuned for code and agent loops."},{"id":"glm-5.3","provider":"Z.ai","type":"chat","modality":"text","context_length":1310720,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":1.75,"output_usdc_per_mtok":5.5,"provider_name":"Z.ai","description":"Flagship agentic coding model, 1.3M context."},{"id":"glm-5.3-flash","provider":"Z.ai","type":"chat","modality":"text","context_length":1310720,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":0.188,"output_usdc_per_mtok":0.625,"provider_name":"Z.ai","description":"The cheap, fast GLM for high-volume agents."},{"id":"deepseek-v4-pro","provider":"DeepSeek","type":"chat","modality":"text","context_length":1048576,"released_at":"2026-08-13","min_output_tokens":512,"input_usdc_per_mtok":1.65,"output_usdc_per_mtok":4.95,"provider_name":"DeepSeek","description":"DeepSeek's strongest open model."},{"id":"deepseek-v4-flash","provider":"DeepSeek","type":"chat","modality":"text","context_length":1310720,"released_at":"2026-07-31","min_output_tokens":512,"input_usdc_per_mtok":0.55,"output_usdc_per_mtok":1.65,"provider_name":"DeepSeek","description":"Fast DeepSeek for loops and retries."},{"id":"gpt-oss-120b","provider":"OpenAI","type":"chat","modality":"text","context_length":128000,"released_at":"2025-08-05","min_output_tokens":512,"input_usdc_per_mtok":0.438,"output_usdc_per_mtok":0.938,"provider_name":"OpenAI","description":"OpenAI's open-weight reasoning model."},{"id":"gpt-oss-20b","provider":"OpenAI","type":"chat","modality":"text","context_length":128000,"released_at":"2025-08-05","min_output_tokens":512,"input_usdc_per_mtok":0.25,"output_usdc_per_mtok":0.375,"provider_name":"OpenAI","description":"Small, quick open-weight reasoner."},{"id":"nemotron-3-120b","provider":"NVIDIA","type":"chat","modality":"text","context_length":256000,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":0.625,"output_usdc_per_mtok":1.88,"provider_name":"NVIDIA","description":"120B MoE with 12B active, built for agents."},{"id":"qwen3.8-27b","provider":"Qwen","type":"chat","modality":"text+vision","context_length":262144,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":0.563,"output_usdc_per_mtok":4,"provider_name":"Qwen","description":"Dense multimodal Qwen, Apache 2.0."},{"id":"qwen3-30b-a3b","provider":"Qwen","type":"chat","modality":"text","context_length":32768,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":0.0637,"output_usdc_per_mtok":0.419,"provider_name":"Qwen","description":"3B active parameters. Pennies per job."},{"id":"qwq-32b","provider":"Qwen","type":"chat","modality":"text","context_length":24000,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":0.825,"output_usdc_per_mtok":1.25,"provider_name":"Qwen","description":"Qwen's reasoning specialist."},{"id":"qwen2.5-coder-32b","provider":"Qwen","type":"chat","modality":"text","context_length":32768,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.825,"output_usdc_per_mtok":1.25,"provider_name":"Qwen","description":"Code completion and repair."},{"id":"glm-4.7-flash","provider":"Z.ai","type":"chat","modality":"text","context_length":131072,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":0.0757,"output_usdc_per_mtok":0.5,"provider_name":"Z.ai","description":"Routing, tagging, summaries at scale."},{"id":"gemma-4-26b","provider":"Google","type":"chat","modality":"text+vision","context_length":256000,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":0.125,"output_usdc_per_mtok":0.375,"provider_name":"Google","description":"Gemma 4 MoE, reads images."},{"id":"llama-3.3-70b-instruct","provider":"Meta","type":"chat","modality":"text","context_length":24000,"released_at":"2024-12-06","min_output_tokens":1,"input_usdc_per_mtok":0.367,"output_usdc_per_mtok":2.82,"provider_name":"Meta","description":"The dependable 70B workhorse."},{"id":"llama-4-scout-17b","provider":"Meta","type":"chat","modality":"text+vision","context_length":131000,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.338,"output_usdc_per_mtok":1.07,"provider_name":"Meta","description":"Llama 4 MoE, multimodal."},{"id":"llama-3.1-8b","provider":"Meta","type":"chat","modality":"text","context_length":32000,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.19,"output_usdc_per_mtok":0.359,"provider_name":"Meta","description":"Small and fast."},{"id":"llama-3.2-3b","provider":"Meta","type":"chat","modality":"text","context_length":80000,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.0637,"output_usdc_per_mtok":0.419,"provider_name":"Meta","description":"Tiny model for classification."},{"id":"mistral-small-3.1-24b","provider":"Mistral","type":"chat","modality":"text","context_length":128000,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.439,"output_usdc_per_mtok":0.694,"provider_name":"Mistral","description":"Balanced 24B with tool calls."},{"id":"deepseek-r1-distill-32b","provider":"DeepSeek","type":"chat","modality":"text","context_length":80000,"released_at":null,"min_output_tokens":512,"input_usdc_per_mtok":0.622,"output_usdc_per_mtok":6.11,"provider_name":"DeepSeek","description":"R1 reasoning distilled into Qwen 32B."},{"id":"granite-4.0-micro","provider":"IBM","type":"chat","modality":"text","context_length":131000,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.0213,"output_usdc_per_mtok":0.14,"provider_name":"IBM","description":"The cheapest tokens in the catalog."},{"id":"sea-lion-v4-27b","provider":"AI Singapore","type":"chat","modality":"text","context_length":128000,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.439,"output_usdc_per_mtok":0.694,"provider_name":"AI Singapore","description":"Southeast Asian languages, incl. Bahasa."},{"id":"bge-m3","provider":"BAAI","type":"embeddings","modality":"text","context_length":60000,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.0148,"output_usdc_per_mtok":0,"provider_name":"BAAI","description":"Multilingual embeddings, 1024 dims."},{"id":"qwen3-embedding-0.6b","provider":"Qwen","type":"embeddings","modality":"text","context_length":8192,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.0148,"output_usdc_per_mtok":0,"provider_name":"Qwen","description":"Compact Qwen embeddings."},{"id":"bge-base-en-v1.5","provider":"BAAI","type":"embeddings","modality":"text","context_length":512,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.0833,"output_usdc_per_mtok":0,"provider_name":"BAAI","description":"English embeddings, 768 dims."},{"id":"bge-large-en-v1.5","provider":"BAAI","type":"embeddings","modality":"text","context_length":512,"released_at":null,"min_output_tokens":1,"input_usdc_per_mtok":0.255,"output_usdc_per_mtok":0,"provider_name":"BAAI","description":"English embeddings, 1024 dims."}]}