Together AI
DeepSeek V4 Pro (Together) token cost
DeepSeek V4 Pro on Together serverless — host comparison vs Fireworks and first-party.
Published list rates
Input
$1.74
per 1M tokens
Output
$3.48
per 1M tokens
Cached input
$0.2
per 1M tokens
Context window
512,000
tokens
Pricing source: Together AI 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
Three views of DeepSeek V4 Pro (Together): sending only, receiving only, and a 70/30 mix. Same list rates, bigger rows.
| Tokens | Input only | Output only | Mix (70% in / 30% out) |
|---|---|---|---|
| 1 | $0.000002 | $0.000003 | $0.000002 |
| 10 | $0.000017 | $0.000035 | $0.000023 |
| 100 | $0.000174 | $0.000348 | $0.000226 |
| 200 | $0.000348 | $0.000696 | $0.000452 |
| 300 | $0.000522 | $0.001044 | $0.000679 |
| 400 | $0.000696 | $0.001392 | $0.000905 |
| 500 | $0.00087 | $0.00174 | $0.001131 |
| 1,000 | $0.00174 | $0.00348 | $0.002262 |
| 2,000 | $0.00348 | $0.00696 | $0.004524 |
| 5,000 | $0.0087 | $0.0174 | $0.01131 |
| 10,000 | $0.0174 | $0.0348 | $0.02262 |
| 25,000 | $0.0435 | $0.087 | $0.05655 |
| 50,000 | $0.087 | $0.174 | $0.1131 |
| 100,000 | $0.174 | $0.348 | $0.2262 |
| 250,000 | $0.435 | $0.87 | $0.5655 |
| 500,000 | $0.87 | $1.74 | $1.131 |
| 1,000,000 | $1.74 | $3.48 | $2.262 |
| 5,000,000 | $8.70 | $17.40 | $11.31 |
| 10,000,000 | $17.40 | $34.80 | $22.62 |
| 50,000,000 | $87.00 | $174.00 | $113.10 |
| 100,000,000 | $174.00 | $348.00 | $226.20 |
| 500,000,000 | $870.00 | $1,740.00 | $1,131.00 |
| 1,000,000,000 | $1,740.00 | $3,480.00 | $2,262.00 |
| 10,000,000,000 | $17,400.00 | $34,800.00 | $22,620.00 |
| 100,000,000,000 | $174,000.00 | $348,000.00 | $226,200.00 |
Tokens for a fixed budget
Tokens DeepSeek V4 Pro (Together) can process for a set budget — send-only, reply-only, and a 70/30 mix.
| Budget | Input tokens | Output tokens | Mixed tokens (70/30) |
|---|---|---|---|
| $0.01 | 5,747 | 2,874 | 4,421 |
| $0.1 | 57,471 | 28,736 | 44,209 |
| $0.5 | 287,356 | 143,678 | 221,043 |
| $1.00 | 574,713 | 287,356 | 442,087 |
| $5.00 | 2.87M | 1.44M | 2.21M |
| $10.00 | 5.75M | 2.87M | 4.42M |
| $20.00 | 11.5M | 5.75M | 8.84M |
| $25.00 | 14.4M | 7.18M | 11.1M |
| $30.00 | 17.2M | 8.62M | 13.3M |
| $50.00 | 28.7M | 14.4M | 22.1M |
| $100.00 | 57.5M | 28.7M | 44.2M |
| $250.00 | 143.7M | 71.8M | 110.5M |
| $500.00 | 287.4M | 143.7M | 221M |
| $1,000.00 | 574.7M | 287.4M | 442.1M |
| $5,000.00 | 2.87B | 1.44B | 2.21B |
| $10,000.00 | 5.75B | 2.87B | 4.42B |
At these rates
- A workload with 70% input and 30% output costs $2.262 per 1 million total tokens and $226.20 per 100 million.
- At the same token count, output costs 2 times the input rate. Response length therefore deserves its own budget limit.
- Filling the 512,000-token context window with input alone would cost about $0.89088, before any output.
- Cached input is $0.2 per million, 89% 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.