Xiaomi: MiMo-V2.6-Flash
mimo-v2.6-flash- Input / Output
- $0.14 / $0.28
- Context
- 1M
- Max output
- 128K
- Modalities
- VisionChat
- Capabilities
- Tool usePrompt Cache
About MiMo-V2.6-Flash
MiMo-V2.6-Flash is the faster, lighter model of Xiaomi's V2.6 generation, built for multimodal coding agents, tool use, and computer operation. Xiaomi reports it surpasses the earlier MiMo-V2.5-Pro, and that its DeepSWE v1.1 score rose from 48.8 to 65.7 over the previous Flash. It is the economical choice in the V2.6 pair.
Where it works well
- Xiaomi's release note reports DeepSWE v1.1 rising from 48.8 to 65.7 against the previous Flash, a gain of about 17 points.
- The release note says it fully surpasses MiMo-V2.5-Pro, so it replaces a flagship-class model at lighter cost.
- V2.6 targets computer-use agents, 3D spatial reasoning, and multimodal perception, and Flash accepts images with text.
- Training used about 750,000 reinforcement-learning trajectories across coding, agent, and vision tasks.
When to choose another model
- Xiaomi positions MiMo-V2.6-Pro as the flagship, so it remains the pick for the hardest agent tasks.
- Output length is capped well below the context window, so very long generated files should be split across turns.
Getting started
Create API key
Create a key in Console, then use it with every model on the platform.
Send your first request
Copy the example for your language and run it against the endpoint.
Text to textResponses APIPOST/v1/responsesAPI format:curl https://api.tokenlab.sh/v1/responses \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-xxx" \ -d '{ "model": "mimo-v2.6-flash", "input": "Hello!" }'
Pricing
The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.
Default spec
per 1M tokens- Official price
- Input $0.14 / Output $0.28 / Cache read $0.0028
- Verified price
- —
- Discount
- —
Cache read
- Official price
- $0.0028
- Verified price
- —
- Discount
- —
| Official priceper 1M tokens | Verified priceper 1M tokens | Discount | |
|---|---|---|---|
| Default spec | Input $0.14 / Output $0.28 / Cache read $0.0028 | — | — |
| Prompt cache pricing | |||
| Cache read | $0.0028 | — | — |
Usage & activity
Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.
Usage & availability
Last 24 hours- Requests
- Success rate
- P95 latency
- Total tokens
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open MiMo-V2.6-Flash in Console with a prompt ready to edit or send.
Help me try mimo-v2.6-flash with a short message at /v1/responses. Show the reply, latency, and cost.
Use cases
Best for- Vision
Everyday coding agent
Drive an editor or terminal agent for bug fixes and small features where response time and cost per task decide adoption.
Screenshot-driven automation
Feed UI screenshots to the model so it can describe a screen and choose the next click or form entry.
High-volume tool routing
Classify incoming requests and call the right internal tools, with image attachments when users send them.
Prompt examples
This screenshot shows the checkout page with a layout bug. Find the likely CSS cause and write the fix.
Fix the failing unit test in this diff, run the suite using the shell tool, and report what changed.
Read the attached invoice image and call the create_expense tool with vendor, date, and total.
This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.
FAQ
What exactly is MiMo-V2.6-Flash?
It is the lighter model in Xiaomi's MiMo-V2.6 release, tuned for coding agents, tool use, and computer operation. Xiaomi positions it as faster and cheaper than V2.6-Pro while beating the earlier V2.5-Pro.
How much better is it than the previous Flash?
Xiaomi reports its DeepSWE v1.1 score moved from 48.8 to 65.7, and agent task pass rates improved by about a quarter in relative terms. These are the maker's own figures.
Flash or Pro in V2.6?
Start with Flash for routine coding and tool-calling traffic. Move to MiMo-V2.6-Pro when a task needs the strongest reasoning, since Pro scored 72.6 against Flash's 65.7 on Xiaomi's DeepSWE v1.1 results.
Can MiMo-V2.6-Flash read images?
Yes, it accepts images with text, so it can analyze screenshots, diagrams, and documents. It returns text, and its thinking mode can be enabled for harder questions.
How much does MiMo-V2.6-Flash cost?
On TokenLab, MiMo-V2.6-Flash costs Input $0.14 / Output $0.28 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.
What are the context window and output limit of MiMo-V2.6-Flash?
MiMo-V2.6-Flash accepts up to 1,048,576 tokens of context and returns up to 131,072 tokens in one response.
Which endpoint should MiMo-V2.6-Flash use?
Use https://api.tokenlab.sh/v1/responses for MiMo-V2.6-Flash. The request example below shows the matching code shape.
Which operations does MiMo-V2.6-Flash support?
MiMo-V2.6-Flash supports Text to text. Select an operation above to see its endpoint and request example.
Compare MiMo-V2.6-Flash
Sources
Reviewed Oct 2, 2026
