# NovAI model catalog (markdown mirror)

111 live models, one OpenAI-compatible endpoint. This file mirrors https://aiapi-pro.com/model-marketplace.html for agents and text-only clients.

## Source of truth

- Live list with ids and context windows: `https://aiapi-pro.com/v1/models`
- Per-model prices (input / output per 1M tokens, or per image/second): `https://aiapi-pro.com/china-ai-leaderboard.json`

Do not hardcode prices from this page; the JSON is regenerated daily.

## Families

| Family | Provider | Representative models |
|---|---|---|
| DeepSeek | DeepSeek | deepseek-v4-pro, deepseek-v4-flash, deepseek-v4-flash-vision |
| Qwen | Alibaba Cloud | qwen3.8-max, qwen3.6-max, qwen-omni-turbo, qwen-image-3.0 |
| GLM | Zhipu | glm-5.3, glm-5.2, glm-4.7-flash (free), glm-4.6v-flash (free) |
| Kimi | Moonshot | kimi-k3, kimi-k2.6, kimi-k2.7-code |
| Doubao / Seedance | ByteDance | doubao-seed-2.0-pro, doubao-seed-2.0-lite, seedance-2.5 |
| MiniMax | MiniMax | minimax-m3, minimax-hailuo-video, minimax-speech-2.6-hd |
| Hunyuan | Tencent | hunyuan-turbos-vision-video, hy3-preview |
| Kling | Kuaishou | kling-v3-std, kling-v3-pro |
| Vidu / Tripo / StepFun | various | vidu-q2, tripo-p1, step-3 |

## Free-to-call models (no credit charge)

`glm-4.7-flash`, `glm-4.6v-flash`, `glm-4.1v-thinking-flash`, `cogview-3-flash` (image), `cogvideox-flash` (video)

Free tier limits: 10 free-model calls/day, 5 requests/minute; 5 free video generations/day. Resets 00:00 UTC+8.

## Pricing model

- 0% platform markup; the per-token rate equals the provider list price (multiplier 1.0)
- Charged balance never expires; the $2 signup credit expires after 14 days
- Byte/vertex pricing verified against provider price sheets; see `/provenance.json` for the billing breakdown