No exchange rates. No plans.
Pay for what you use.
Credit never expires. Top up from ¥10. Every request is calculated to six decimals.
Model catalog
The table shows list prices and what you pay for each model, including applicable discounts and time bands.
| Model | Pricing tier | Input / 1M tokens | Cached input | Output / 1M tokens | Context |
|---|---|---|---|---|---|
| DeepSeek V4 Pro 0813Reasoning · 1M context · 50% off list | Peak (01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-08-20 13:03:49.577 UTC; peak rates apply all day before then | 1M | |||
| Off-peak (outside 01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-08-20 13:03:49.577 UTC; peak rates apply all day before then | 1M | ||||
| DeepSeek V4.1 FlashReasoning · 1M context · 50% off list | Off-peak (outside 01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-08-16 16:00:00.000 UTC; peak rates apply all day before then | 1M | |||
| Peak (01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-08-16 16:00:00.000 UTC; peak rates apply all day before then | 1M | ||||
| Qwen4 MaxReasoning · 1M context · 15% off list | Standard | 1M | |||
| GLM-5.3 FlashReasoning · 1M context · 50% off list | Standard | 1M | |||
| Qwen3.8 MaxChat · 991K context · 50% off list | Standard | 991K | |||
| GLM-5.3Reasoning · 1.024M context · 50% off list | Standard | 1.024M | |||
| Kimi K3Reasoning · 1.048576M context · 20% off list | Standard | 1.048576M | |||
| Kimi K3.1 Pro HighspeedReasoning · 1.048576M context · 50% off list | Standard | 1.048576M | |||
| Qwen3.8 FlashChat · 1M context · 50% off list | Standard | 1M | |||
| Qwen3.8 27BChat · 1M context · 70% off list | Standard | 1M | |||
| Qwen3.8 2.4T A95BReasoning · 1M context · 20% off list | Standard | 1M | |||
| Qwen3.7 MaxChat · 991K context · 50% off list | Standard | 991K | |||
| Qwen3.6 PlusChat · 1M context · 20% off list | Input ≤ 256,000 tokens | Regular input rate | 1M | ||
| Input ≥ 256,001 tokens | Regular input rate | 1M | |||
| Qwen3.7 PlusChat · 1M context · 20% off list | Input ≤ 256,000 tokens | 1M | |||
| Input ≥ 256,001 tokens | 1M | ||||
| DeepSeek V4.1 Flash 1017Chat · 1.024M context · 50% off list | Off-peak (outside 01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-10-18 23:42:09.074 UTC; peak rates apply all day before then | 1.024M | |||
| Peak (01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-10-18 23:42:09.074 UTC; peak rates apply all day before then | 1.024M |
DeepSeek V4 Pro 0813
Reasoning · 1M context · 50% off list
- Input / 1M tokens
List ¥9¥4.5- Output / 1M tokens
List ¥27¥13.5
Pricing tier · Cached input · Context
Peak (01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-08-20 13:03:49.577 UTC; peak rates apply all day before then
- Cached input
List ¥0.3¥0.15- Context
- 1M
- Input / 1M tokens
List ¥4.5¥2.25- Output / 1M tokens
List ¥13.5¥6.75
Pricing tier · Cached input · Context
Off-peak (outside 01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-08-20 13:03:49.577 UTC; peak rates apply all day before then
- Cached input
List ¥0.15¥0.075- Context
- 1M
DeepSeek V4.1 Flash
Reasoning · 1M context · 50% off list
- Input / 1M tokens
List ¥1¥0.5- Output / 1M tokens
List ¥4¥2
Pricing tier · Cached input · Context
Off-peak (outside 01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-08-16 16:00:00.000 UTC; peak rates apply all day before then
- Cached input
List ¥0.02¥0.01- Context
- 1M
- Input / 1M tokens
List ¥2¥1- Output / 1M tokens
List ¥8¥4
Pricing tier · Cached input · Context
Peak (01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-08-16 16:00:00.000 UTC; peak rates apply all day before then
- Cached input
List ¥0.04¥0.02- Context
- 1M
Qwen4 Max
Reasoning · 1M context · 15% off list
- Input / 1M tokens
List ¥10¥8.5- Output / 1M tokens
List ¥30¥25.5
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥0.5¥0.425- Context
- 1M
GLM-5.3 Flash
Reasoning · 1M context · 50% off list
- Input / 1M tokens
List ¥0.8¥0.4- Output / 1M tokens
List ¥2.8¥1.4
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥0.23¥0.115- Context
- 1M
Qwen3.8 Max
Chat · 991K context · 50% off list
- Input / 1M tokens
List ¥12¥6- Output / 1M tokens
List ¥36¥18
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥1.5¥0.75- Context
- 991K
GLM-5.3
Reasoning · 1.024M context · 50% off list
- Input / 1M tokens
List ¥8¥4- Output / 1M tokens
List ¥28¥14
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥2¥1- Context
- 1.024M
Kimi K3
Reasoning · 1.048576M context · 20% off list
- Input / 1M tokens
List ¥20¥16- Output / 1M tokens
List ¥100¥80
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥2¥1.6- Context
- 1.048576M
Kimi K3.1 Pro Highspeed
Reasoning · 1.048576M context · 50% off list
- Input / 1M tokens
List ¥26¥13- Output / 1M tokens
List ¥136¥68
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥1.3¥0.65- Context
- 1.048576M
Qwen3.8 Flash
Chat · 1M context · 50% off list
- Input / 1M tokens
List ¥0.8¥0.4- Output / 1M tokens
List ¥2.7¥1.35
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥0.1¥0.05- Context
- 1M
Qwen3.8 27B
Chat · 1M context · 70% off list
- Input / 1M tokens
List ¥3¥0.9- Output / 1M tokens
List ¥12¥3.6
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥0.6¥0.18- Context
- 1M
Qwen3.8 2.4T A95B
Reasoning · 1M context · 20% off list
- Input / 1M tokens
List ¥12¥9.6- Output / 1M tokens
List ¥36¥28.8
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥1.5¥1.2- Context
- 1M
Qwen3.7 Max
Chat · 991K context · 50% off list
- Input / 1M tokens
List ¥12¥6- Output / 1M tokens
List ¥36¥18
Pricing tier · Cached input · Context
Standard
- Cached input
List ¥2.4¥1.2- Context
- 991K
Qwen3.6 Plus
Chat · 1M context · 20% off list
- Input / 1M tokens
List ¥2¥1.6- Output / 1M tokens
List ¥12¥9.6
Pricing tier · Cached input · Context
Input ≤ 256,000 tokens
- Cached input
- Regular input rate
- Context
- 1M
- Input / 1M tokens
List ¥8¥6.4- Output / 1M tokens
List ¥48¥38.4
Pricing tier · Cached input · Context
Input ≥ 256,001 tokens
- Cached input
- Regular input rate
- Context
- 1M
Qwen3.7 Plus
Chat · 1M context · 20% off list
- Input / 1M tokens
List ¥2¥1.6- Output / 1M tokens
List ¥8¥6.4
Pricing tier · Cached input · Context
Input ≤ 256,000 tokens
- Cached input
List ¥0.4¥0.32- Context
- 1M
- Input / 1M tokens
List ¥6¥4.8- Output / 1M tokens
List ¥24¥19.2
Pricing tier · Cached input · Context
Input ≥ 256,001 tokens
- Cached input
List ¥1.2¥0.96- Context
- 1M
DeepSeek V4.1 Flash 1017
Chat · 1.024M context · 50% off list
- Input / 1M tokens
List ¥1¥0.5- Output / 1M tokens
List ¥4¥2
Pricing tier · Cached input · Context
Off-peak (outside 01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-10-18 23:42:09.074 UTC; peak rates apply all day before then
- Cached input
List ¥0.1¥0.05- Context
- 1.024M
- Input / 1M tokens
List ¥2¥1- Output / 1M tokens
List ¥8¥4
Pricing tier · Cached input · Context
Peak (01:00–04:00, 06:00–10:00 UTC (Beijing time 09:00–12:00, 14:00–18:00)) · Time bands from 2026-10-18 23:42:09.074 UTC; peak rates apply all day before then
- Cached input
List ¥0.2¥0.1- Context
- 1.024M
Image models
Rates are CNY per image, using each model’s current discount. Reference and generated images are charged separately. See the table for resolution tiers and prompt-enhancement rates; the regular output rate applies when no enhancement rate is listed.
| Model | Reference / image | Output / image | Enhanced output / image |
|---|---|---|---|
| Qwen Image 3.0 ProImage generation · per image · 20% off list | 262,144–2,250,000 pixels | — | |
| Qwen Image 3.0Image generation · per image · 20% off list | 262,144–4,194,304 pixels | — | |
| Wan 2.7 Image ProImage generation · per image · 20% off list | ¥0 | 589,824–16,777,216 pixels | — |
| Wan 2.7 ImageImage generation · per image · 20% off list | ¥0 | 589,824–4,194,304 pixels | — |
| Z-Image TurboImage generation · per image · 20% off list | ¥0 | 262,144–4,194,304 pixels | 262,144–4,194,304 pixels |
Qwen Image 3.0 Pro
Image generation · per image · 20% off list
- Reference / image
List ¥0.02¥0.016- Output / image
- 262,144–2,250,000 pixels
List ¥0.25¥0.22,250,001–4,194,304 pixelsList ¥0.5¥0.4
Enhanced output / image
—Qwen Image 3.0
Image generation · per image · 20% off list
- Reference / image
List ¥0.02¥0.016- Output / image
- 262,144–4,194,304 pixels
List ¥0.18¥0.144
Enhanced output / image
—Wan 2.7 Image Pro
Image generation · per image · 20% off list
- Reference / image
- ¥0
- Output / image
- 589,824–16,777,216 pixels
List ¥0.5¥0.4
Enhanced output / image
—Wan 2.7 Image
Image generation · per image · 20% off list
- Reference / image
- ¥0
- Output / image
- 589,824–4,194,304 pixels
List ¥0.2¥0.16
Enhanced output / image
—Z-Image Turbo
Image generation · per image · 20% off list
- Reference / image
- ¥0
- Output / image
- 262,144–4,194,304 pixels
List ¥0.1¥0.08
Enhanced output / image
262,144–4,194,304 pixelsExplicit cache rates
Temporary upstream promotions are excluded; discounts use official list prices. Only the explicit cache creation and hit rates available here are listed.
Each request reserves its estimated cost (at least ¥0.10) from the available account and key budgets. Running requests and pending charges also use that budget. If it is insufficient, wait for requests to finish or lower the output or separate thinking budget.
Each account can run up to 4 model requests at once. A reservation is not a charge; actual usage determines the final bill. Usage above the estimate is still charged and may make the balance negative. Top-ups first cover any amount owed. Without an output limit, the model maximum is reserved. Separate thinking budgets are added; unspecified thinking budgets conservatively use the context limit.
Time bands and input-length ranges are configured per model; see the UTC and Beijing-time windows and tier conditions above. Compare selling rates for the same input range rather than assuming a fixed discount or half-price relationship.
Input-token count and request start time select the pricing tier. Settlement uses the active CNY platform price for that request.
Comparing flagship models? See live Kimi K3, Qwen3.8 Flash and GLM-5.3 Flash rates →
Corporate purchasing, invoicing and purchase orders → support@kimiseek.app
Once per account with no successful top-ups. The first ¥500 of your first top-up is 30% off; the rest is full price. Save up to ¥150. Unused discount allowance does not carry over.
Choose an amount
For trials and personal projects
For reliable small applications
A stronger starting balance for most projects