> ## Documentation Index
> Fetch the complete documentation index at: https://liaobots.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.
# Models & Pricing
> Liaobots is committed to providing developers with fast, convenient, and affordable API interface solutions, building a stable and easy-to-use API platform that integrates almost all AI large models in one place.
## Billing Rules
**Billing Method**: Billing follows OpenAI API's token-based pricing method, where 1000 tokens ≈ 500-750 words. All prices in the table below are per **1 million (1M) tokens**.
**Cache Billing Method**: The "Cache Read/Create" column consistently uses the main-site and standard OpenAI-compatible API credit rates, shown as `read/create per 1M`. Cache reads use a model-specific configured price when available; otherwise they fall back to point input price × 0.1. Standard endpoints do not charge for cache creation, shown as `0`; fixed-price models (image/video) do not use caching, shown as `-`. Claude Code (`/v1/messages`) uses a separate official-ratio pricing path described under "Claude Code Billing Method" below and does not use the prices in this column.
**Gemini 3.6/3.7 Flash promotion**: The prices shown are valid through **December 31, 2026** and will be adjusted to Google's standard pricing starting January 1, 2027.
**Tiered Billing Method**: Some models are priced in tiers based on input length. For those models the "Point Price", "Official Price (USD)" and "Cache Read/Create" columns below list every tier as `≤272K: … ; >272K: …`, and the model list on the main site shows a "Tiered billing" badge. The tier is selected by the **total input tokens of a single request** (including cached tokens) and excludes output tokens. The upper bound is inclusive: exactly 272K stays on the standard tier, while 272K + 1 moves to the long-context tier. The whole request is billed at the matched tier — it is never split and summed across tiers. The GPT-5.6 family (Sol / Terra / Luna) uses tiered billing today.
**Claude Code Billing Method**: The billing algorithm is consistent with the official, with a dynamically adjusted price ratio. Current ratio: 1 USD = 3.5 credits. Cache reads are billed at official input price × current ratio × 0.1; 5-minute cache creation is × 1.25 and 1-hour cache creation is × 2. The `-t` discounts in the Point Price column apply only to the main site and standard API, not to Claude Code.
**CodeX Billing Method**: The billing algorithm is consistent with the official, with dynamically adjusted price ratio. Current ratio: 1 USD = 0.5 credits
**Credits to RMB Ratio**: Credits and RMB are at a 1:1 ratio. In this article, you can see the ratio between our prices and official prices.
**Voice reading billing** (main-site features, not API tokens):
| Feature | Pricing |
| ------------------------------------------------------------ | ---------------------------------------------------------------------------------- |
| Multi-voice / streaming multi-voice / custom voice synthesis | By text length: **0.25 credits per 500 characters** (partial blocks count as full) |
| Single-voice reading | Per generation: **0.05 credits** |
**Long Conversation Billing Principle**: The principle of long conversations with AI is that each conversation implicitly sends all previous content to the AI. As conversations get longer, the cumulative content sent also increases, so token consumption will also increase.
## How to Integrate
We support API calls for various models. No format adaptation is needed. Use OpenAI API format, just modify base\_url and api\_key. api\_key uses the main site Login Code, pay-as-you-go.
OpenAI Official API Documentation
**base\_url**: `https://ai.liaobots.work/v1` (for backup URLs, see [API Call Examples](/Getting-Started/en/API))
**api\_key**: Login Code
```shell theme={null}
curl --location 'https://ai.liaobots.work/v1/chat/completions' \
--header 'Authorization: Bearer Login Code' \
--header 'Content-Type: application/json' \
--data '{
"model": "gpt-4o",
"stream": false,
"messages": [
{
"role": "user",
"content": "hi"
}
]
}'
```
## Model List
The following prices are for **1M tokens (1 million tokens)**. Official prices are in USD, and credits can be simply calculated as RMB. Latest exchange rate (\~7.3): [usd-to-cny-rate](https://wise.com/zh-cn/currency-converter/usd-to-cny-rate)
Prices may change with market fluctuations, but we will always insist on providing lower prices than official
| Provider | Model Name | Context | Point Price | Cache Read/Create | Price Ratio | Official Price (USD) |
| ----------------- | --------------------------------------- | ------- | ---------------------------------------------------------------------- | ------------------------------------- | ----------- | ------------------------------------------------------------------ |
| Anthropic | claude-sonnet-5-t | 1M | Input 3/M, Output 15/M | 0.3/0 per 1M | 20% | Input 2/M, Output 10/M |
| Anthropic | claude-sonnet-5 | 1M | Input 7/M, Output 35/M | 0.7/0 per 1M | 47% | Input 2/M, Output 10/M |
| Anthropic | claude-sonnet-4-6 | 1M | Input 5.25/M, Output 26.25/M | 0.525/0 per 1M | 47% | Input 1.5/M, Output 7.5/M |
| Anthropic | claude-sonnet-4-5-20250929-thinking | 200K | Input 5.25/M, Output 26.25/M | 0.525/0 per 1M | 47% | Input 1.5/M, Output 7.5/M |
| Anthropic | claude-sonnet-4-5-20250929-t | 200K | Input 2.25/M, Output 11.25/M | 0.225/0 per 1M | 20% | Input 1.5/M, Output 7.5/M |
| Anthropic | claude-sonnet-4-5-20250929 | 200K | Input 5.25/M, Output 26.25/M | 0.525/0 per 1M | 47% | Input 1.5/M, Output 7.5/M |
| Anthropic | claude-opus-5-t | 1M | Input 7.5/M, Output 37.5/M | 0.75/0 per 1M | 20% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-5 | 1M | Input 17.5/M, Output 87.5/M | 1.75/0 per 1M | 47% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-4-8 | 1M | Input 17.5/M, Output 87.5/M | 1.75/0 per 1M | 47% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-4-7-t | 1M | Input 7.5/M, Output 37.5/M | 0.75/0 per 1M | 20% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-4-7 | 1M | Input 17.5/M, Output 87.5/M | 1.75/0 per 1M | 47% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-4-6-thinking | 1M | Input 17.5/M, Output 87.5/M | 1.75/0 per 1M | 47% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-4-6-t | 1M | Input 7.5/M, Output 37.5/M | 0.75/0 per 1M | 20% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-4-6 | 1M | Input 17.5/M, Output 87.5/M | 1.75/0 per 1M | 47% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-4-5-20251101-t | 200K | Input 7.5/M, Output 37.5/M | 0.75/0 per 1M | 20% | Input 5/M, Output 25/M |
| Anthropic | claude-opus-4-1-20250805-t | 200K | Input 22.5/M, Output 112.5/M | 2.25/0 per 1M | 20% | Input 15/M, Output 75/M |
| Anthropic | claude-opus-4-20250514 | 200K | Input 52.5/M, Output 262.5/M | 5.25/0 per 1M | 47% | Input 15/M, Output 75/M |
| Anthropic | claude-fable-5-1 | 1M | Input 35/M, Output 175/M | 3.5/0 per 1M | 47% | Input 10/M, Output 50/M |
| Anthropic | claude-fable-5 | 1M | Input 35/M, Output 175/M | 3.5/0 per 1M | 47% | Input 10/M, Output 50/M |
| chat-server-video | veo-3.1-standard | 0K | Input 0.8, Output 0 | - | - | Input 0, Output 0 |
| chat-server-video | veo-3.1-fast | 0K | Input 0.6, Output 0 | - | - | Input 0, Output 0 |
| chat-server-video | seedance-2-5 | 0K | Input 0, Output 0 | - | - | Input 0, Output 0 |
| chat-server-video | seedance-2-0-fast | 0K | Input 0, Output 0 | - | - | Input 0, Output 0 |
| chat-server-video | seedance-2-0 | 0K | Input 0, Output 0 | - | - | Input 0, Output 0 |
| chat-server-video | grok-imagine-video-1.5-preview | 0K | Input 0, Output 0 | - | - | Input 0, Output 0 |
| DeepSeek | deepseek-v4.1-flash | 1M | Input 1.1/M, Output 4.38/M | 0.022/0 per 1M | 50% | Input 0.3/M, Output 1.2/M |
| DeepSeek | deepseek-v4-pro-0813 | 1M | Input 4.82/M, Output 14.45/M | 0.16/0 per 1M | 49% | Input 1.32/M, Output 3.96/M |
| DeepSeek | deepseek-v4-pro | 1M | Input 4.82/M, Output 14.45/M | 0.16/0 per 1M | 49% | Input 1.32/M, Output 3.96/M |
| DeepSeek | deepseek-v4-flash-0731 | 1M | Input 1.61/M, Output 4.82/M | 0.05/0 per 1M | 50% | Input 0.44/M, Output 1.32/M |
| DeepSeek | deepseek-v4-flash-preview | 1M | Input 1.61/M, Output 4.82/M | 0.05/0 per 1M | 50% | Input 0.44/M, Output 1.32/M |
| DeepSeek | deepseek-v4-flash | 1M | Input 1.61/M, Output 4.82/M | 0.05/0 per 1M | 50% | Input 0.44/M, Output 1.32/M |
| DeepSeek | deepseek-v3.2 | 128K | Input 1.568/M, Output 2.352/M | 0.1568/0 per 1M | 76% | Input 0.28/M, Output 0.42/M |
| GLM | glm-5.3-flash | 1M | Input 0.2/M, Output 0.56/M | 0.046/0 per 1M | 15% | Input 0.15/M, Output 0.5/M |
| GLM | glm-5.3 | 1M | Input 2/M, Output 5.6/M | 0.2/0 per 1M | 17% | Input 1.4/M, Output 4.4/M |
| GLM | glm-5.2-fast | 1M | Input 3/M, Output 8.4/M | 0.3/0 per 1M | 17% | Input 2.1/M, Output 6.6/M |
| GLM | glm-5.2 | 1M | Input 2/M, Output 5.6/M | 0.2/0 per 1M | 17% | Input 1.4/M, Output 4.4/M |
| GLM | glm-5.1 | 200K | Input 2/M, Output 5.6/M | 0.2/0 per 1M | 17% | Input 1.4/M, Output 4.4/M |
| GLM | glm-4.7 | 200K | Input 1.6/M, Output 7/M | 0.16/0 per 1M | 54% | Input 0.4/M, Output 1.75/M |
| Google | gemini-3.1-flash-image-preview | 200K | Input 0.2, Output 0.2 | - | 13% | Input 0.2, Output 0.2 |
| Google | gemini-3-pro-image-preview-openrouter | 256K | Input 0.2, Output 0.2 | - | 6% | Input 0.4, Output 0.4 |
| Google | gemini-3.1-flash-lite-image | 200K | Input 0.12, Output 0.12 | - | 5% | Input 0.3, Output 0.3 |
| Google | gemini-3-pro-image-preview | 256K | Input 0.25, Output 0.25 | - | 8% | Input 0.4, Output 0.4 |
| Google | gemini-2.5-flash-image | 32K | Input 0.1, Output 0.1 | - | 6% | Input 0.2, Output 0.2 |
| Google | imagen-4.0-ultra-generate-preview-06-06 | 100K | Input 0.2, Output 0.2 | - | 2% | Input 1, Output 1 |
| Google | imagen-4.0-generate-preview-06-06 | 100K | Input 0.1, Output 0.1 | - | 1% | Input 1, Output 1 |
| Google | gemini-3.8-flash | 1M | Input 1.875/M, Output 9.375/M | 0.1875/0 per 1M | 34% | Input 0.75/M, Output 3.75/M |
| Google | gemini-3.7-flash | 1M | Input 1.875/M, Output 9.375/M | 0.1875/0 per 1M | 34% | Input 0.75/M, Output 3.75/M |
| Google | gemini-3.6-flash | 1M | Input 1.875/M, Output 9.375/M | 0.1875/0 per 1M | 34% | Input 0.75/M, Output 3.75/M |
| Google | gemini-3.5-flash | 1M | Input 3.75/M, Output 22.5/M | 0.375/0 per 1M | 34% | Input 1.5/M, Output 9/M |
| Google | gemini-3.1-pro-preview-vertex | 1M | Input 3.2/M, Output 19.2/M | 0.32/0 per 1M | 21% | Input 2/M, Output 12/M |
| Google | gemini-3.1-pro-preview | 1M | Input 5/M, Output 30/M | 0.5/0 per 1M | 34% | Input 2/M, Output 12/M |
| Google | gemini-3.1-flash-lite-preview | 1M | Input 0.625/M, Output 3.75/M | 0.0625/0 per 1M | 34% | Input 0.25/M, Output 1.5/M |
| Google | gemini-3-pro-preview-vertex | 1M | Input 3.2/M, Output 19.2/M | 0.32/0 per 1M | 21% | Input 2/M, Output 12/M |
| Google | gemini-3-pro-preview-thinking | 1M | Input 5/M, Output 30/M | 0.5/0 per 1M | 34% | Input 2/M, Output 12/M |
| Google | gemini-3-pro-preview | 1M | Input 5/M, Output 30/M | 0.5/0 per 1M | 34% | Input 2/M, Output 12/M |
| Google | gemini-3-flash-preview | 1M | Input 1.25/M, Output 7.5/M | 0.125/0 per 1M | 34% | Input 0.5/M, Output 3/M |
| Google | gemini-2.5-pro-vertex | 1M | Input 2/M, Output 16/M | 0.2/0 per 1M | 21% | Input 1.25/M, Output 10/M |
| Google | gemini-2.5-pro | 1M | Input 3.125/M, Output 25/M | 0.3125/0 per 1M | 34% | Input 1.25/M, Output 10/M |
| Google | gemini-2.5-flash | 1M | Input 0.75/M, Output 6.25/M | 0.075/0 per 1M | 34% | Input 0.3/M, Output 2.5/M |
| Kimi | kimi-k3 | 1M | Input 16.8/M, Output 84/M | 1.68/0 per 1M | 76% | Input 3/M, Output 15/M |
| Kimi | kimi-k2.7-code | 262K | Input 1.9/M, Output 8/M | 0.19/0 per 1M | 27% | Input 0.95/M, Output 4/M |
| Kimi | kimi-k2.6 | 262K | Input 1.9/M, Output 8/M | 0.19/0 per 1M | 27% | Input 0.95/M, Output 4/M |
| Kimi | kimi-k2.5 | 262K | Input 1.8/M, Output 8.8/M | 0.18/0 per 1M | 54% | Input 0.45/M, Output 2.2/M |
| Kimi | kimi-k2-thinking | 256K | Input 2.4/M, Output 10/M | 0.24/0 per 1M | 54% | Input 0.6/M, Output 2.5/M |
| MiniMax | minimax-m3 | 1M | Input 2.4/M, Output 9.6/M | 0.24/0 per 1M | 54% | Input 0.6/M, Output 2.4/M |
| MiniMax | minimax-m2.7 | 205K | Input 1.2/M, Output 4.8/M | 0.12/0 per 1M | 54% | Input 0.3/M, Output 1.2/M |
| MiniMax | minimax-m2.5 | 205K | Input 1.2/M, Output 4.8/M | 0.12/0 per 1M | 54% | Input 0.3/M, Output 1.2/M |
| OpenAI | o4-mini | 200K | Input 1.1/M, Output 4.4/M | 0.11/0 per 1M | 13% | Input 1.1/M, Output 4.4/M |
| OpenAI | o3-mini | 200K | Input 1.1/M, Output 4.4/M | 0.11/0 per 1M | 13% | Input 1.1/M, Output 4.4/M |
| OpenAI | o3 | 200K | Input 4/M, Output 16/M | 0.4/0 per 1M | 27% | Input 2/M, Output 8/M |
| OpenAI | horizon-beta | 256K | Input 0.01, Output 0.01 | - | 1% | Input 0.1, Output 0.1 |
| OpenAI | gpt-image-2 | 128K | Input 0.15, Output 0.15 | - | 13% | Input 0.15, Output 0.15 |
| OpenAI | gpt-6-astra | 1.05M | ≤272K: Input 8/M, Output 40/M; >272K: Input 16/M, Output 60/M | ≤272K: 0.8/0; >272K: 1.6/0 per 1M | 10% | ≤272K: Input 10/M, Output 50/M; >272K: Input 20/M, Output 75/M |
| OpenAI | gpt-5.6-terra | 1.05M | ≤272K: Input 1.6/M, Output 9.6/M; >272K: Input 3.2/M, Output 19.2/M | ≤272K: 0.16/0; >272K: 0.32/0 per 1M | 10% | ≤272K: Input 2/M, Output 12/M; >272K: Input 4/M, Output 24/M |
| OpenAI | gpt-5.6-sol | 1.05M | ≤272K: Input 4/M, Output 24/M; >272K: Input 8/M, Output 36/M | ≤272K: 0.4/0; >272K: 0.8/0 per 1M | 10% | ≤272K: Input 5/M, Output 30/M; >272K: Input 10/M, Output 45/M |
| OpenAI | gpt-5.6-luna | 1.05M | ≤272K: Input 0.16/M, Output 0.96/M; >272K: Input 0.32/M, Output 1.44/M | ≤272K: 0.016/0; >272K: 0.032/0 per 1M | 10% | ≤272K: Input 0.2/M, Output 1.2/M; >272K: Input 0.4/M, Output 1.8/M |
| OpenAI | gpt-5.5 | 1.05M | Input 10/M, Output 60/M | 1/0 per 1M | 27% | Input 5/M, Output 30/M |
| OpenAI | gpt-5.4-nano | 400K | Input 0.4/M, Output 2.5/M | 0.04/0 per 1M | 27% | Input 0.2/M, Output 1.25/M |
| OpenAI | gpt-5.4-mini | 400K | Input 1.5/M, Output 9/M | 0.15/0 per 1M | 27% | Input 0.75/M, Output 4.5/M |
| OpenAI | gpt-5.4 | 1.05M | Input 5/M, Output 30/M | 0.5/0 per 1M | 27% | Input 2.5/M, Output 15/M |
| OpenAI | gpt-5.2 | 400K | Input 3.5/M, Output 28/M | 0.35/0 per 1M | 27% | Input 1.75/M, Output 14/M |
| OpenAI | gpt-5-codex | 400K | Input 2.5/M, Output 20/M | 0.25/0 per 1M | 27% | Input 1.25/M, Output 10/M |
| OpenAI | gpt-5-chat-latest | 128K | Input 4/M, Output 16/M | 0.4/0 per 1M | 27% | Input 2/M, Output 8/M |
| OpenAI | gpt-5-mini | 400K | Input 0.5/M, Output 4/M | 0.05/0 per 1M | 27% | Input 0.25/M, Output 2/M |
| OpenAI | gpt-5 | 400K | Input 2.5/M, Output 20/M | 0.25/0 per 1M | 27% | Input 1.25/M, Output 10/M |
| OpenAI | gpt-4o-mini-free | 8K | Input 0.01/M, Output 0.01/M | 0.001/0 per 1M | 0% | Input 0.15/M, Output 0.6/M |
| OpenAI | gpt-4o-mini-2024-07-18 | 128K | Input 0.3/M, Output 1.2/M | 0.03/0 per 1M | 27% | Input 0.15/M, Output 0.6/M |
| OpenAI | gpt-4o-mini | 128K | Input 0.3/M, Output 1.2/M | 0.03/0 per 1M | 27% | Input 0.15/M, Output 0.6/M |
| OpenAI | gpt-4o-free | 8K | Input 0.01/M, Output 0.01/M | 0.001/0 per 1M | 13% | Input 0.01/M, Output 0.01/M |
| OpenAI | gpt-4o-2024-11-20 | 128K | Input 5/M, Output 20/M | 0.5/0 per 1M | 27% | Input 2.5/M, Output 10/M |
| OpenAI | gpt-4o-2024-08-06 | 128K | Input 5/M, Output 20/M | 0.5/0 per 1M | 27% | Input 2.5/M, Output 10/M |
| OpenAI | gpt-4o-2024-05-13 | 128K | Input 10/M, Output 30/M | 1/0 per 1M | 27% | Input 5/M, Output 15/M |
| OpenAI | gpt-4o-2024-11-20 | 128K | Input 5/M, Output 20/M | 0.5/0 per 1M | 27% | Input 2.5/M, Output 10/M |
| OpenAI | gpt-4.1-nano | 1M | Input 0.7/M, Output 2.8/M | 0.07/0 per 1M | 95% | Input 0.1/M, Output 0.4/M |
| OpenAI | gpt-4.1-nano-2025-04-14 | 1M | Input 0.7/M, Output 2.8/M | 0.07/0 per 1M | 95% | Input 0.1/M, Output 0.4/M |
| OpenAI | gpt-4.1-mini | 1M | Input 2.8/M, Output 11.2/M | 0.28/0 per 1M | 95% | Input 0.4/M, Output 1.6/M |
| OpenAI | gpt-4.1-mini-2025-04-14 | 1M | Input 2.8/M, Output 11.2/M | 0.28/0 per 1M | 95% | Input 0.4/M, Output 1.6/M |
| OpenAI | gpt-4.1-2025-04-14 | 1M | Input 4/M, Output 16/M | 0.4/0 per 1M | 27% | Input 2/M, Output 8/M |
| OpenAI | gpt-4.1 | 1M | Input 4/M, Output 16/M | 0.4/0 per 1M | 27% | Input 2/M, Output 8/M |
| OpenAI | gpt-4-turbo-2024-04-09 | 128K | Input 20/M, Output 60/M | 2/0 per 1M | 27% | Input 10/M, Output 30/M |
| OpenAI | gpt-4-turbo | 128K | Input 20/M, Output 60/M | 2/0 per 1M | 27% | Input 10/M, Output 30/M |
| OpenAI | gpt-image-2.5-sunburst | - | Input 0.15, Output 0.15 | - | - | Input 0, Output 0 |
| OpenAI | gpt-image-2.5-flare | - | Input 0.15, Output 0.15 | - | - | Input 0, Output 0 |
| qwen | qwen3.7-max | 1M | Input 9.125/M, Output 27.375/M | 0.9125/0 per 1M | 50% | Input 2.5/M, Output 7.5/M |
| qwen | qwen3-max-2026-01-23 | 262K | Input 4/M, Output 16/M | 0.4/0 per 1M | 99% | Input 0.55/M, Output 2.21/M |
| x.ai | grok-4-6 | 500K | Input 2/M, Output 6/M | 0.5/0 per 1M | 13% | Input 2/M, Output 6/M |
| x.ai | grok-4-5 | 256K | Input 2/M, Output 6/M | 0.2/0 per 1M | 13% | Input 2/M, Output 6/M |
| x.ai | grok-4-fast | 2M | Input 0.2/M, Output 0.5/M | 0.02/0 per 1M | 13% | Input 0.2/M, Output 0.5/M |
| x.ai | grok-3 | 128K | Input 1.2/M, Output 6/M | 0.12/0 per 1M | 5% | Input 3/M, Output 15/M |
| xAI | grok-imagine-image-quality | 128K | Input 0.1, Output 0.1 | - | 13% | Input 0.1, Output 0.1 |
| xAI | grok-imagine-image | 128K | Input 0.06, Output 0.06 | - | 13% | Input 0.06, Output 0.06 |