Microsoft
MAI-Thinking-1 token cost
Microsoft's first reasoning model (1T total / 35B active MoE) at $2/$8 per 1M with a 256K context window.
Published list rates
Input
$2.00
per 1M tokens
Output
$8.00
per 1M tokens
Context window
256,000
tokens
Pricing source: Microsoft Foundry MAI 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
MAI-Thinking-1 at rising volume: prompt-only, reply-only, and a mixed chat (70% in, 30% out).
| Tokens | Input only | Output only | Mix (70% in / 30% out) |
|---|---|---|---|
| 1 | $0.000002 | $0.000008 | $0.000004 |
| 10 | $0.00002 | $0.00008 | $0.000038 |
| 100 | $0.0002 | $0.0008 | $0.00038 |
| 200 | $0.0004 | $0.0016 | $0.00076 |
| 300 | $0.0006 | $0.0024 | $0.00114 |
| 400 | $0.0008 | $0.0032 | $0.00152 |
| 500 | $0.001 | $0.004 | $0.0019 |
| 1,000 | $0.002 | $0.008 | $0.0038 |
| 2,000 | $0.004 | $0.016 | $0.0076 |
| 5,000 | $0.01 | $0.04 | $0.019 |
| 10,000 | $0.02 | $0.08 | $0.038 |
| 25,000 | $0.05 | $0.2 | $0.095 |
| 50,000 | $0.1 | $0.4 | $0.19 |
| 100,000 | $0.2 | $0.8 | $0.38 |
| 250,000 | $0.5 | $2.00 | $0.95 |
| 500,000 | $1.00 | $4.00 | $1.90 |
| 1,000,000 | $2.00 | $8.00 | $3.80 |
| 5,000,000 | $10.00 | $40.00 | $19.00 |
| 10,000,000 | $20.00 | $80.00 | $38.00 |
| 50,000,000 | $100.00 | $400.00 | $190.00 |
| 100,000,000 | $200.00 | $800.00 | $380.00 |
| 500,000,000 | $1,000.00 | $4,000.00 | $1,900.00 |
| 1,000,000,000 | $2,000.00 | $8,000.00 | $3,800.00 |
| 10,000,000,000 | $20,000.00 | $80,000.00 | $38,000.00 |
| 100,000,000,000 | $200,000.00 | $800,000.00 | $380,000.00 |
Tokens for a fixed budget
If you cap spend, this is roughly how many MAI-Thinking-1 tokens that money buys.
| Budget | Input tokens | Output tokens | Mixed tokens (70/30) |
|---|---|---|---|
| $0.01 | 5,000 | 1,250 | 2,632 |
| $0.1 | 50,000 | 12,500 | 26,316 |
| $0.5 | 250,000 | 62,500 | 131,579 |
| $1.00 | 500,000 | 125,000 | 263,158 |
| $5.00 | 2.5M | 625,000 | 1.32M |
| $10.00 | 5M | 1.25M | 2.63M |
| $20.00 | 10M | 2.5M | 5.26M |
| $25.00 | 12.5M | 3.13M | 6.58M |
| $30.00 | 15M | 3.75M | 7.89M |
| $50.00 | 25M | 6.25M | 13.2M |
| $100.00 | 50M | 12.5M | 26.3M |
| $250.00 | 125M | 31.3M | 65.8M |
| $500.00 | 250M | 62.5M | 131.6M |
| $1,000.00 | 500M | 125M | 263.2M |
| $5,000.00 | 2.5B | 625M | 1.32B |
| $10,000.00 | 5B | 1.25B | 2.63B |
At these rates
- A workload with 70% input and 30% output costs $3.80 per 1 million total tokens and $380.00 per 100 million.
- At the same token count, output costs 4 times the input rate. Response length therefore deserves its own budget limit.
- Filling the 256,000-token context window with input alone would cost about $0.512, before any output.
Public preview on Microsoft Foundry; rates are the published starting prices for input and output per 1M tokens.
These figures use public list prices. Cache eligibility, batch or priority modes, long-context surcharges, discounts, and regional billing can change the invoice.