One agent session
1M input tokens and 100K output tokens, no caching.
- Input1M x $0.435 = $0.435
- Output0.1M x $0.87 = $0.087
$0.522per session
Xiaomi MiMo - Open weights
MiMo-V2.6-Pro pricing and specs
MiMo-V2.6-Pro is an open-weights model from Xiaomi MiMo that costs $0.435 per 1M input tokens and $0.87 per 1M output tokens on the first-party API, with a 1M-token context window.
Last verified from mimo.mi.com. All sources
- Input / 1M
- $0.435
- Output / 1M
- $0.87
- Context
- 1M tokens
- Weights
- Open
Pricing per 1M tokens
| Item | Price / 1M tokens |
|---|---|
| Input | $0.435 |
| Output | $0.87 |
| Cache read | $0.0036 |
| Batch inputBatch requests | $0.2175 |
| Batch outputBatch requests | $0.435 |
Note: Overseas USD prices (the China region is billed separately in RMB). Cache writes are listed as limited-time free, so cache write is not stated. Batch API also lists a cache-hit input price of $0.0018 per 1M (no schema field). The model page states context 1M tokens and max output 128K tokens; max output is recorded as 131072 because Xiaomi's own docs use that exact figure for the V2.6 series (mimo-v2.6-flash default max_completion_tokens 131072 on the deprecation page, UltraSpeed sample code max_completion_tokens=131072). context window 1000000 follows the stated 1M (the gateway lists 1048576). Input: text, image, video, audio; output: text. mimo-v2.5-pro requests auto-route here since 2026-10-14 and that id stops working 2026-10-21. A separate higher-priced UltraSpeed variant is sold as mimo-v2.6-pro-ultraspeed (its own entry). Weights are the MiMo-V2.6-Pro-RL checkpoint (1.02T total / 42B active); the HF card names the Xiaomi MiMo Open Platform API as a place it is served. A Token Plan subscription is sold separately and is not reflected here.
Specs
- Context window
- 1,000,000 tokens
- Max output
- 131,072 tokens
- Release date
- Provider
- Xiaomi MiMo
- API model id
mimo-v2.6-pro- Vercel AI Gateway id
xiaomi/mimo-v2.6-pro- Weights
- Open weights
- License
- MIT (Hugging Face license tag: mit)
- Weights download
- huggingface.co
What MiMo-V2.6-Pro costs in practice
Worked from the standard list rates above, with the math shown so you can swap in your own token counts.
A chat app
10,000 requests a day at 2K input and 500 output tokens each.
- Input20M x $0.435 = $8.70
- Output5M x $0.87 = $4.35
$13.05per day
$391.50 per 30-day month
The same session at a 90% cache hit
90% of the input is read from the prompt cache. The one-time cache write premium is not included.
- Cached input0.9M x $0.0036 = $0.0032
- Uncached input0.1M x $0.435 = $0.0435
- Output0.1M x $0.87 = $0.087
$0.1337per session
Compare with
| Model | Input / 1M | Output / 1M |
|---|---|---|
| MiMo-V2.6-Pro (this page) | $0.435 | $0.87 |
| Codestral 25.08Mistral AI | $0.30 | $0.90 |
| GLM-4.6VZ.ai | $0.30 | $0.90 |
| Mercury 2Inception | $0.25 | $0.75 |
| Mercury 2.5Inception | $0.20 | $0.75 |
| MiMo-V2.6-FlashXiaomi MiMo | $0.14 | $0.28 |
FAQ
How much does MiMo-V2.6-Pro cost per million tokens?
MiMo-V2.6-Pro costs $0.435 per 1M input tokens and $0.87 per 1M output tokens on Xiaomi MiMo's first-party API, per mimo.mi.com on October 17, 2026. Per 1M tokens, cache reads cost $0.0036, and the batch rate is $0.2175 input and $0.435 output.
What is MiMo-V2.6-Pro's context window?
MiMo-V2.6-Pro has a 1,000,000-token context window and returns up to 131,072 output tokens per response.
Are MiMo-V2.6-Pro's weights open?
Yes. MiMo-V2.6-Pro's weights are published at https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL under the MIT (Hugging Face license tag: mit).
When was MiMo-V2.6-Pro released?
Xiaomi MiMo released MiMo-V2.6-Pro on September 22, 2026.
Sources
Last verified from mimo.mi.com.
Newsletter
Get model price changes as they ship
One email per week. New model releases, price cuts and what they mean for your bill.