基础套餐
适合日常使用的轻量月度额度。
US$19.90 / 月
每月到账 US$25.00 额度
透明的零加价定价
ModelMax 不在推理成本上加价。充值余额后，按实际模型用量扣费。
月度额度订阅
每月付款并获得更多 ModelMax 额度，额度会累加到余额且永久有效。
热门
高级套餐
为活跃项目提供更多月度额度。
US$59.90 / 月
每月到账 US$80.00 额度
旗舰套餐
面向高负载使用的最高月度额度。
US$159.90 / 月
每月到账 US$250.00 额度
说明
Google Gemini
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| gemini-3.1-pro-preview | $2.00 | $12.00 | Input LC: $4.00 | Output LC: $18.00 |
| gemini-3-pro-preview | $2.00 | $12.00 | Input LC: $4.00 | Output LC: $18.00 |
| gemini-3.1-flash-image-preview | $0.50 | $3.00 | Image: $60/1M pixels |
| gemini-3-flash-preview | $0.50 | $3.00 | Audio in: $1.00/1M |
| gemini-3.1-flash-lite-preview | $0.25 | $1.50 | Audio in: $0.50/1M |
Google Veo (Video)
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| veo-3.1 | $0.20/s | $0.40/s | 4K: $0.40/s | Audio: +$0.20/s |
| veo-3.1-fast | $0.10/s | $0.15/s | 4K: $0.30/s | Audio: +$0.05/s |
| veo-3 | $0.20/s | $0.40/s | Audio: +$0.20/s |
| veo-3-fast | $0.10/s | $0.15/s | Audio: +$0.05/s |
| veo-2 | $0.50/s | - | Standard Gen |
xAI Grok on Google Cloud
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| grok-4.3 | $1.25 | $2.50 | Cache: $0.20 | LC input: $2.50 | LC output: $5.00 |
| grok-4.1-fast-non-reasoning | $0.20 | $0.50 | Cache: $0.05 |
OpenAI
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| gpt-5.5 | $5.00 | $30.00 | Flagship GPT |
| gpt-5.4 | $2.50 | $15.00 | High capability |
| gpt-5.4-mini | $0.75 | $4.50 | Balanced |
| gpt-5.4-nano | $0.20 | $1.25 | Low latency |
Anthropic Claude
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| claude-opus-4-8 | $5.00 | $25.00 | Highest capability |
| claude-sonnet-4-6 | $3.00 | $15.00 | Balanced intelligence |
| claude-haiku-4-5 | $1.00 | $5.00 | Fast and efficient |
DeepSeek
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| deepseek-r1 | $1.35 | $5.40 | Reasoning with Chain of Thought |
| deepseek-v3.1 | $0.58 | $1.68 | General Purpose |
| deepseek-v3.2 | $0.62 | $1.85 | General Purpose (Latest) |
Qwen
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| qwen3-coder-30b-a3b | $0.15 | $0.60 | Code-focused |
| qwen3-32b | $0.15 | $0.60 | General Purpose |
| qwen3-235b-a22b-2507 | $0.22 | $0.88 | Large Frontier |
| qwen3-coder-480b-a35b | $0.45 | $1.80 | Large Code |
| qwen3-next-80b-a3b | $0.15 | $1.20 | Next Gen MoE |
| qwen3-vl-235b-a22b | $0.53 | $2.66 | Vision-Language |
| qwen3-coder-next | $0.50 | $1.20 | Latest Code |
Kimi
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| kimi-k2-thinking | $0.60 | $2.50 | Extended Thinking |
| kimi-k2.5 | $0.60 | $3.00 | General Purpose |
MiniMax
| 模型 | 输入 | 输出 | 说明 |
|---|---|---|---|
| minimax-m2 | $0.30 | $1.20 | General Purpose |
| minimax-m2.1 | $0.30 | $1.20 | General Purpose |
Note: "LC" refers to Long Context tiered pricing. Caching read prices are typically heavily discounted on supported models (e.g. Gemini, Anthropic). Video models are billed per second of generated video length, with variations based on resolution (4K vs 1080p) and audio generation.
打开控制台充值
ModelMax 不在推理成本上加价。充值余额后，按实际模型用量扣费。