SiliconFlow
Qwen3.5 9B (SiliconFlow) token cost calculator
Qwen3.5 9B on SiliconFlow for low-cost classification and high-volume chat.
Published list rates
Input
$0.1
per 1M tokens
Output
$0.15
per 1M tokens
Context window
262,000
tokens
Pricing source: SiliconFlow pricing
Token cost calculator
Estimate from text
Calculate from a token total
Project monthly volume
Uses the input/output mix above and a 30-day month: requests per day × tokens per request × 30.
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 by token volume
Compare input-only, output-only, and 70/30 mixed workloads from one token to 100 billion. Intermediate bands make smaller production steps easier to price.
| Tokens | Input only | Output only | 70/30 mix |
|---|---|---|---|
| 1 | $0.0000001 | $0.00000015 | $0.000000115 |
| 10 | $0.000001 | $0.000002 | $0.000001 |
| 100 | $0.00001 | $0.000015 | $0.000012 |
| 200 | $0.00002 | $0.00003 | $0.000023 |
| 300 | $0.00003 | $0.000045 | $0.000035 |
| 400 | $0.00004 | $0.00006 | $0.000046 |
| 500 | $0.00005 | $0.000075 | $0.000058 |
| 1,000 | $0.0001 | $0.00015 | $0.000115 |
| 2,000 | $0.0002 | $0.0003 | $0.00023 |
| 5,000 | $0.0005 | $0.00075 | $0.000575 |
| 10,000 | $0.001 | $0.0015 | $0.00115 |
| 25,000 | $0.0025 | $0.00375 | $0.002875 |
| 50,000 | $0.005 | $0.0075 | $0.00575 |
| 100,000 | $0.01 | $0.015 | $0.0115 |
| 250,000 | $0.025 | $0.0375 | $0.02875 |
| 500,000 | $0.05 | $0.075 | $0.0575 |
| 1,000,000 | $0.1 | $0.15 | $0.115 |
| 5,000,000 | $0.5 | $0.75 | $0.575 |
| 10,000,000 | $1.00 | $1.50 | $1.15 |
| 50,000,000 | $5.00 | $7.50 | $5.75 |
| 100,000,000 | $10.00 | $15.00 | $11.50 |
| 500,000,000 | $50.00 | $75.00 | $57.50 |
| 1,000,000,000 | $100.00 | $150.00 | $115.00 |
| 10,000,000,000 | $1,000.00 | $1,500.00 | $1,150.00 |
| 100,000,000,000 | $10,000.00 | $15,000.00 | $11,500.00 |
Tokens available by budget
See how many tokens a fixed budget buys at the published rates. Display-currency values use the current exchange-rate snapshot shown on the page.
| Budget | Input tokens | Output tokens | Mixed tokens (70/30) |
|---|---|---|---|
| $0.01 | 100,000 | 66,667 | 86,957 |
| $0.1 | 1M | 666,667 | 869,565 |
| $0.5 | 5M | 3.33M | 4.35M |
| $1.00 | 10M | 6.67M | 8.7M |
| $5.00 | 50M | 33.3M | 43.5M |
| $10.00 | 100M | 66.7M | 87M |
| $20.00 | 200M | 133.3M | 173.9M |
| $25.00 | 250M | 166.7M | 217.4M |
| $30.00 | 300M | 200M | 260.9M |
| $50.00 | 500M | 333.3M | 434.8M |
| $100.00 | 1B | 666.7M | 869.6M |
| $250.00 | 2.5B | 1.67B | 2.17B |
| $500.00 | 5B | 3.33B | 4.35B |
| $1,000.00 | 10B | 6.67B | 8.7B |
| $5,000.00 | 50B | 33.3B | 43.5B |
| $10,000.00 | 100B | 66.7B | 87B |
Cost profile
- A workload with 70% input and 30% output costs $0.115 per 1 million total tokens and $11.50 per 100 million.
- At the same token count, output costs 1.5 times the input rate. Response length therefore deserves its own budget limit.
- Filling the 262,000-token context window with input alone would cost about $0.0262, before any output.
How the text estimate works
For rough planning, CostUse uses about four characters per token and about 0.75 words per token. Actual tokenization changes with language, model, formatting, and special tokens.
These results use public list prices. Cache eligibility, batch or priority modes, long-context surcharges, discounts, and regional billing can change the final invoice.
Questions about this rate
How much is 1 million mixed tokens on Qwen3.5 9B (SiliconFlow)?
About $0.115 at a 70% input / 30% output mix using list prices verified 2026-07-30.
How much are 100 million mixed tokens?
About $11.50 at the same 70/30 mix — useful for rough monthly volume planning.
What is the Qwen3.5 9B (SiliconFlow) price per million input and output tokens?
Input $0.1 and output $0.15 per 1M tokens (USD list rates). Use the calculator above with your real token mix.
Is the text-to-token estimate exact?
No. The estimator approximates tokens from characters and words. Real bills depend on the provider tokenizer, special tokens, and discounts (cache, batch, priority).
What does cost per token mean, and why is output usually more expensive?
Providers charge for tokens processed. Output tokens (generated replies) often cost several times more than input tokens (your prompt), so completion length can dominate the bill.