OpenAI-compatible API

Ship the model layer
in minutes.

TokensHub gives your application one predictable endpoint for prepaid AI credits. Start with the request format your team already knows.

Quick start

After creating an account and an API key, send a standard chat completion request.

curl https://tokenshub.com.au/v1/chat/completions \ -H "Authorization: Bearer YOUR_TOKENSHUB_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-v4-flash", "messages": [{"role": "user", "content": "Hello"}], "max_tokens": 80 }'
Use GET /v1/models to list the exact model IDs currently enabled. Preview models are excluded.

Model routing

Pick a model by capability. TokensHub handles the prepaid credit settlement behind the same API shape.

deepseek-v4.1-flash-1017deepseek-v4.1-flashdeepseek-v4-flashDeepSeek-V3.2deepseek-v4-flash-0731glm-5.3-flashqwen3.8-flashqwen3.6-flashqwen-flash-charactermimo-v2.6-flashMiniMax-M3.1-Flashhy3MiniMax-M2.5deepseek-v4-prodeepseek-v4-flash-vision-expmimo-v2.6-proMiniMax-M3MiniMax-M2.7glm-5kimi-k2.5qwen3.6-35b-a3bqwen3.6-flash-2026-04-16qwen3.6-plusqwen3.7-plusglm-5.2glm-5.3glm-5-turboglm-5.1Doubao-Seed-2.1-prokimi-k2.6kimi-k2.7-codeqwen3.6-27bqwen3.6-plus-2026-04-02qwen3.7-maxqwen3.7-max-2026-05-20qwen3.8-maxdeepseek-v4-pro-0813hy4-previewkimi-k3.1-prokimi-k3qwen3.8-max-0902

Video & Image endpoints

Use the same Bearer API key and prepaid credit balance across video and image generation endpoints. Text to Video accepts wan3.0-video (480p, 720p or 1080p; optional generate_audio:true) and wan3.0-video-prime (720p or 1080p, silent). Standard supports 2, 5, 10 or 15 seconds; Prime supports 2, 5 or 10 seconds. TokensHub pins the dtld and sl7 routes respectively. Frame to Video is available in the signed-in T Motion workspace. The website and API share daily and active-video limits.

# 1. Video generation (Wan 3.0 — requires Idempotency-Key header) curl https://tokenshub.com.au/v1/video/generations \ -H "Authorization: Bearer YOUR_TOKENSHUB_API_KEY" \ -H "Idempotency-Key: wan-job-000000000001" \ -H "Content-Type: application/json" \ -d '{"model":"wan3.0-video","prompt":"A cobalt coat passes amber-lit windows","aspect_ratio":"16:9","resolution":"720p","duration":5,"generate_audio":false}' # Keep the same Idempotency-Key for retries; use a new key for a new video. # 2. Poll video status (replace TASK_ID with the ID returned above) curl https://tokenshub.com.au/v1/video/generations/TASK_ID \ -H "Authorization: Bearer YOUR_TOKENSHUB_API_KEY" # 3. Image generation (Qwen Image 3.0 / Wan 2.7 Image / Z Image Turbo) curl https://tokenshub.com.au/v1/images/generations \ -H "Authorization: Bearer YOUR_TOKENSHUB_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"qwen-image-3.0","prompt":"A minimalist sapphire token emblem","size":"1024x1024","n":1}'

Credits and spend

Pay as you go with prepaid credits. Each request consumes credits at the published model rate. See your balance and usage in the dashboard. One-time top-ups are available in your dashboard.