{"info":{"name":"Tokens API","schema":"https://schema.getpostman.com/json/collection/v2.1.0/collection.json","description":"Set the baseUrl and apiKey variables. Docs: https://tokens.bd/docs"},"variable":[{"key":"baseUrl","value":"https://tokens.bd"},{"key":"apiKey","value":""}],"item":[{"name":"POST /v1/chat/completions","request":{"method":"POST","header":[{"key":"Authorization","value":"Bearer {{apiKey}}"},{"key":"Content-Type","value":"application/json"}],"url":{"raw":"{{baseUrl}}/v1/chat/completions","host":["{{baseUrl}}"],"path":["v1","chat","completions"]},"body":{"mode":"raw","raw":"{\n  \"model\": \"deepseek/deepseek-v4.1-flash\",\n  \"messages\": [\n    {\n      \"role\": \"user\",\n      \"content\": \"What does HTTP 429 mean?\"\n    }\n  ],\n  \"max_tokens\": 300\n}"},"description":"OpenAI-style chat. The body is forwarded to the model's provider; fields a model does not support are rejected by it."}},{"name":"POST /v1/completions","request":{"method":"POST","header":[{"key":"Authorization","value":"Bearer {{apiKey}}"},{"key":"Content-Type","value":"application/json"}],"url":{"raw":"{{baseUrl}}/v1/completions","host":["{{baseUrl}}"],"path":["v1","completions"]},"body":{"mode":"raw","raw":"{\n  \"model\": \"deepseek/deepseek-v4.1-flash\",\n  \"prompt\": \"Say hello\",\n  \"max_tokens\": 20\n}"},"description":"Prompt in, text out. Works only for models whose provider still serves it."}},{"name":"POST /v1/messages","request":{"method":"POST","header":[{"key":"Authorization","value":"Bearer {{apiKey}}"},{"key":"Content-Type","value":"application/json"}],"url":{"raw":"{{baseUrl}}/v1/messages","host":["{{baseUrl}}"],"path":["v1","messages"]},"body":{"mode":"raw","raw":"{\n  \"model\": \"deepseek/deepseek-v4.1-flash\",\n  \"max_tokens\": 300,\n  \"messages\": [\n    {\n      \"role\": \"user\",\n      \"content\": \"What does HTTP 429 mean?\"\n    }\n  ]\n}"},"description":"Anthropic-style messages. Use the gateway root as the SDK's base URL; the SDKs add /v1/messages."}},{"name":"POST /v1/messages/count_tokens","request":{"method":"POST","header":[{"key":"Authorization","value":"Bearer {{apiKey}}"}],"url":{"raw":"{{baseUrl}}/v1/messages/count_tokens","host":["{{baseUrl}}"],"path":["v1","messages","count_tokens"]},"description":"Not billed. Same body as /v1/messages."}},{"name":"POST /v1/embeddings","request":{"method":"POST","header":[{"key":"Authorization","value":"Bearer {{apiKey}}"},{"key":"Content-Type","value":"application/json"}],"url":{"raw":"{{baseUrl}}/v1/embeddings","host":["{{baseUrl}}"],"path":["v1","embeddings"]},"body":{"mode":"raw","raw":"{\n  \"model\": \"<an embedding model id>\",\n  \"input\": \"text to embed\"\n}"},"description":"Only for catalog models that are embedding models."}},{"name":"POST /v1/responses","request":{"method":"POST","header":[{"key":"Authorization","value":"Bearer {{apiKey}}"},{"key":"Content-Type","value":"application/json"}],"url":{"raw":"{{baseUrl}}/v1/responses","host":["{{baseUrl}}"],"path":["v1","responses"]},"body":{"mode":"raw","raw":"{\n  \"model\": \"deepseek/deepseek-v4.1-flash\",\n  \"input\": \"What does HTTP 429 mean?\"\n}"},"description":"OpenAI's Responses API, for tools such as Codex CLI. Works when the model's upstream supports it."}},{"name":"GET /v1/models","request":{"method":"GET","header":[{"key":"Authorization","value":"Bearer {{apiKey}}"}],"url":{"raw":"{{baseUrl}}/v1/models","host":["{{baseUrl}}"],"path":["v1","models"]},"description":"Filtered by the key's allowed models, your plan and your wallet. Not billed."}},{"name":"GET /v1/tokens/usage","request":{"method":"GET","header":[{"key":"Authorization","value":"Bearer {{apiKey}}"}],"url":{"raw":"{{baseUrl}}/v1/tokens/usage","host":["{{baseUrl}}"],"path":["v1","tokens","usage"]},"description":"Specific to Tokens. Not billed and not counted toward your request limit."}}]}