SiliconFlow
DeepSeek V3.2 (SiliconFlow) token cost
DeepSeek V3.2 on SiliconFlow with cache discounts for production open routing.
Published list rates
Input
$0.27
per 1M tokens
Output
$0.42
per 1M tokens
Cached input
$0.135
per 1M tokens
Context window
164,000
tokens
Pricing source: SiliconFlow pricing
Estimate cost
Estimate from text
Calculate from a token total
Project monthly volume
Daily requests × tokens per request × 30 days.
Estimated cost
- Estimated tokens
- —
- Characters
- —
- Words
- —
- Input tokens
- —
- Output tokens
- —
- Input cost
- —
- Output cost
- —
- Total cost
- —
- Combined price per 1M tokens
- —
- Estimated monthly cost
- —
- Estimated daily cost
- —
Cost as volume grows
What DeepSeek V3.2 (SiliconFlow) costs if you only send tokens, only get a reply, or do both (70% in / 30% out). Rows start small and go up to huge traffic.
| Tokens | Input only | Output only | Mix (70% in / 30% out) |
|---|---|---|---|
| 1 | $0.00000027 | $0.00000042 | $0.000000315 |
| 10 | $0.000003 | $0.000004 | $0.000003 |
| 100 | $0.000027 | $0.000042 | $0.000032 |
| 200 | $0.000054 | $0.000084 | $0.000063 |
| 300 | $0.000081 | $0.000126 | $0.000095 |
| 400 | $0.000108 | $0.000168 | $0.000126 |
| 500 | $0.000135 | $0.00021 | $0.000157 |
| 1,000 | $0.00027 | $0.00042 | $0.000315 |
| 2,000 | $0.00054 | $0.00084 | $0.00063 |
| 5,000 | $0.00135 | $0.0021 | $0.001575 |
| 10,000 | $0.0027 | $0.0042 | $0.00315 |
| 25,000 | $0.00675 | $0.0105 | $0.007875 |
| 50,000 | $0.0135 | $0.021 | $0.01575 |
| 100,000 | $0.027 | $0.042 | $0.0315 |
| 250,000 | $0.0675 | $0.105 | $0.07875 |
| 500,000 | $0.135 | $0.21 | $0.1575 |
| 1,000,000 | $0.27 | $0.42 | $0.315 |
| 5,000,000 | $1.35 | $2.10 | $1.575 |
| 10,000,000 | $2.70 | $4.20 | $3.15 |
| 50,000,000 | $13.50 | $21.00 | $15.75 |
| 100,000,000 | $27.00 | $42.00 | $31.50 |
| 500,000,000 | $135.00 | $210.00 | $157.50 |
| 1,000,000,000 | $270.00 | $420.00 | $315.00 |
| 10,000,000,000 | $2,700.00 | $4,200.00 | $3,150.00 |
| 100,000,000,000 | $27,000.00 | $42,000.00 | $31,500.00 |
Tokens for a fixed budget
If you cap spend, this is roughly how many DeepSeek V3.2 (SiliconFlow) tokens that money buys.
| Budget | Input tokens | Output tokens | Mixed tokens (70/30) |
|---|---|---|---|
| $0.01 | 37,037 | 23,810 | 31,746 |
| $0.1 | 370,370 | 238,095 | 317,460 |
| $0.5 | 1.85M | 1.19M | 1.59M |
| $1.00 | 3.7M | 2.38M | 3.17M |
| $5.00 | 18.5M | 11.9M | 15.9M |
| $10.00 | 37M | 23.8M | 31.7M |
| $20.00 | 74.1M | 47.6M | 63.5M |
| $25.00 | 92.6M | 59.5M | 79.4M |
| $30.00 | 111.1M | 71.4M | 95.2M |
| $50.00 | 185.2M | 119M | 158.7M |
| $100.00 | 370.4M | 238.1M | 317.5M |
| $250.00 | 925.9M | 595.2M | 793.7M |
| $500.00 | 1.85B | 1.19B | 1.59B |
| $1,000.00 | 3.7B | 2.38B | 3.17B |
| $5,000.00 | 18.5B | 11.9B | 15.9B |
| $10,000.00 | 37B | 23.8B | 31.7B |
At these rates
- A workload with 70% input and 30% output costs $0.315 per 1 million total tokens and $31.50 per 100 million.
- At the same token count, output costs 1.56 times the input rate. Response length therefore deserves its own budget limit.
- Filling the 164,000-token context window with input alone would cost about $0.04428, before any output.
- Cached input is $0.135 per million, 50% below standard input when the workload meets the provider’s cache rules.
These figures use public list prices. Cache eligibility, batch or priority modes, long-context surcharges, discounts, and regional billing can change the invoice.