Pricing
This page provides pricing information for Vanchin models and tools. All prices are in USD.
1、Pay as you go
Model list：
KwaiKAT Model | Input Length | Input Price (Million tokens) | Output Price (Million tokens) | Cache Write (Million tokens) | Cache Read (Million tokens) |
|---|---|---|---|---|---|
KAT-Coder-Pro-V2.5 | 0-256K | $0.74 | $2.96 | - | $0.15 |
KAT-Coder-Air-V2.5 | 0-256K | $0.15 | $0.60 | - | $0.03 |
KAT-Coder-Pro-V2 | 0-256K | $0.3 | $1.2 | Free | $0.06 |
KAT-Coder-Pro-V1 | 0-256K | $0.3 | $1.2 | Free | $0.06 |
KAT-Coder-Pro V1 (31%OFF) Limited Offer Period: Jan 5, 2026 - Feb 5, 2026
Qwen Model | Input price (Million tokens) | Cached Input Price (Million tokens) | Output price (Million tokens) |
Qwen3-235B-A22B-Instruct-2507 | $0.350 | $0.070 | $1.400 |
Qwen3-235B-A22B-Thinking-2507 | $0.350 | - | $4.200 |
Qwen3-30B-A3B-Instruct-2507 | $0.107 | $0.022 | $0.429 |
Qwen3-30B-A3B | $0.107 | $0.043 | Thinking：$1.071 NonThinking：$0.429 |
Qwen3-VL-235B-A22B-Instruct | $0.350 | $0.057 | $1.400 |
Qwen3-Coder-480B-A35B-Instruct-FP8 | $0.9 | $0.18 | $4.5 |
Qwen3-30B-A3B-Thinking-2507 | $0.2 | $0.04 | $2.4 |
Qwen3-VL-235B-A22B-Thinking | $0.4 | - | $4.0 |
Qwen3-Coder-Next | $0.3 | $0.06 | $1.5 |
Qwen3.5-397B-A17B | $0.6 | $0.12 | $3.6 |
Qwen3.5-122B-A10B | $0.4 | $0.08 | $3.2 |
Qwen3.5-27B | $0.3 | $0.06 | $2.4 |
Qwen3.5-35B-A3B | $0.25 | $0.05 | $2 |
Qwen3.6-35B-A3B | $0.375 | - | $2.25 |
Qwen3.6-27B | $0.60 | - | $3.60 |
DeepSeek Model | Input price (Million tokens) | Cached Input Price (Million tokens) | Output price (Million tokens) |
DeepSeek-R1-250528 | $0.571 | $0.229 | $2.286 |
DeepSeek-V3 | $0.286 | $0.114 | $1.143 |
DeepSeek-V3.1 | $0.571 | $0.229 | $1.714 |
DeepSeek-V3.2-Exp | $0.286 | $0.0286 | $0.429 |
DeepSeek-V3.2 | $0.286 | $0.0286 | $0.429 |
DeepSeek-V3.2-speciale | $0.286 | $0.0286 | $0.429 |
DeepSeek-V3.1-Terminus | $0.571 | $0.229 | $1.714 |
DeepSeek-V4-Flash-0423 | $0.14 | $0.028 | $0.28 |
DeepSeek-V4-Flash-0731 | $0.14 | $0.028 | $0.28 |
DeepSeek-V4-Pro | $1.74 | $0.145 | $3.48 |
GLM Model | Input price (Million tokens) | Cached Input Price (Million tokens) | Output price (Million tokens) |
GLM-4.6 | $0.60 | $0.11 | $2.20 |
GLM-4.7 | $0.60 | $0.11 | $2.20 |
GLM-5 | $1.00 | $0.20 | $3.20 |
GLM-5.1 | $1.40 | $0.26 | $4.40 |
GLM-5.2 | $1.40 | $0.26 | $4.40 |
Kimi Model | Input price (Million tokens) | Cached Input Price (Million tokens) | Output price (Million tokens) |
Kimi-K2-Instruct-0905 | $0.571 | $0.143 | $2.286 |
Kimi K2.5 | $0.60 | $0.10 | $3.00 |
Kimi K2.6 | $0.95 | $0.16 | $4.00 |
Kimi-K2.7-Code | $0.95 | $0.19 | $4.00 |
MiniMax Model | Input price (Million tokens) | Output price (Million tokens) | Cache Write (Million tokens) | Cache Read (Million tokens) |
MiniMax-M2 | $0.300 | $1.200 | $0.375 | $0.030 |
MiniMax-M2.1 | $0.300 | $1.200 | $0.375 | $0.030 |
MiniMax-M2.5 | $0.300 | $1.200 | $0.375 | $0.030 |
MiMo Model | Input price (Million tokens) | Cached Input Price (Million tokens) | Output price (Million tokens) |
MiMo-V2.5-Pro | $0.435 | $0.0036 | $0.87 |
2、KwaiKAT Coding Plan
Coding Plan for KAT-Coder-Pro V2 → Purchase
3、Prepaid Resource Packs（ Offline from Jan 5, 2026）
Currently, only prepaid resource packages are available for purchase. UnitToken price: $0.6/Million UnitTokens
We provide a free trial resource package of 20M UnitTokens for each new account. If the resource package does not meet your needs, you can contact us for customized purchase.
KAT-Coder-Pro-V1 | Quota | Discount | Price after discount | Term of validity |
Trial Package | 20 Million UnitTokens | Each account is eligible to claim once | $0 | 90 Days |
Package 1 | 34 Million UnitTokens | -2% | $19.9 | 90 Days |
Package 2 | 105 Million UnitTokens | -5% | $59.9 | 90 Days |
Package 3 | 185 Million UnitTokens | -10% | $99.9 | 90 Days |
Package 4 | 392 Million UnitTokens | -15% | $199.9 | 90 Days |
Package 5 | 834 Million UnitTokens | -20% | $399.9 | 90 Days |
Package 6 | 1459 Million UnitTokens | -20% | $699.9 | 90 Days |
Package 7 | 2222 Million UnitTokens | -25% | $999.9 | 90 Days |
Customized Resource Package | Customized | More | Customized | Customized |
Deduction Rules：
1. The deduction unit for resource packages is the number of unit tokens, which varies depending on the length of the input token and token type：
【 0-32K 】 1 input token deducts 1 unit token, 1 output token deducts 4 unit tokens, 1 prefix cache hit deducts 0.2 unit token.
【 32-128K 】 1 input token deducts 1.5 unit tokens, 1 output token deducts 6 unit tokens, 1 prefix cache hit deducts 0.3 unit token.
【 128-256K 】 1 input token deducts 2.5 unit tokens, 1 output token deducts 10 unit tokens, 1 prefix cache hit deducts 0.5 unit token.
2. Unused resource packages will be liquidated upon expiration.
Remarks：No credits will be deducted for failed tasks, regardless of the reason for failure, including failures due to content moderation policies.