Where is the MiMo-V2.6-Flash documentation?

The MiMo-V2.6-Flash documentation on OpenRouter includes the How to Benchmark LLMs blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.

How much does MiMo-V2.6-Flash cost?

MiMo-V2.6-Flash costs $0.10/M input tokens and $0.28/M output tokens, with separate rates for Cache Read at $0.0028/M tokens.

What is the context length of MiMo-V2.6-Flash?

MiMo-V2.6-Flash has a 1,050,000 token context window. It supports up to 131,072 completion tokens.

Does MiMo-V2.6-Flash support tool calling and structured outputs?

Yes. MiMo-V2.6-Flash accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

What inputs and outputs does MiMo-V2.6-Flash support?

MiMo-V2.6-Flash accepts text, images, video, and audio as input and returns text.

Which providers serve MiMo-V2.6-Flash?

MiMo-V2.6-Flash is served by 8 providers on OpenRouter: Darkbloom, GMICloud, Cohere, DeepInfra, Xiaomi, Tenstorrent, Venice, and Makora. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

When was MiMo-V2.6-Flash released?

MiMo-V2.6-Flash was released on September 21, 2026.

Xiaomi: MiMo-V2.6-Flash

xiaomi/mimo-v2.6-flash

Model weights

Compare

MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. Built on a Mixture-of-Experts architecture with 309B total parameters and 15B activated per token, it employs a hybrid attention mechanism for greater computational efficiency. The model features a 1M-token context window and native multimodal capabilities. Optimized for agentic workflows, it delivers strong performance across coding, visual, general, and research scenarios, excelling at complex, long-horizon tasks with robust generalization across a diverse range of agent harnesses.

Modalities

In / out price

$0.10 / $0.28per 1M

Context

1.1M

Released

Sep 21, 2026

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput


	$0.18	$0.36	$0.0045	0.53s	177 tok/s	99.82%
	$0.14	$0.28	$0.003	0.51s	64 tok/s	98.29%
	$0.14	$0.28	$0.0028	3.22s	30 tok/s	100.00%
20% off	$0.14$0.112	$0.28$0.224	$0.003$0.0024	2.79s	28 tok/s	99.95%
	$0.165	$0.33	$0.004	0.72s	24 tok/s	95.29%
	$0.10	$0.28	$0.0028	8.51s	19 tok/s	99.10%
	$0.175	$0.35	$0.00375	3.75s	34 tok/s	89.89%
	$0.14	$0.28	$0.0028	1.82s	35 tok/s	34.90%

Input $0.18Output $0.36
177 tok/s99.82% uptime
Input $0.14Output $0.28
64 tok/s98.29% uptime
Input $0.14Output $0.28
30 tok/s100.00% uptime
Input $0.112Output $0.224
20% off
28 tok/s99.95% uptime
Input $0.165Output $0.33
24 tok/s95.29% uptime
Input $0.10Output $0.28
19 tok/s99.10% uptime
Input $0.175Output $0.35
34 tok/s89.89% uptime
Input $0.14Output $0.28
35 tok/s34.90% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.01601/M tokens

Output

$0.2994/M tokens

MiMo-V2.6-Flash — Price history

Price history

No effective price data for this selection

$0.0130	$0.280	92.6%	89.5%
$0.0194	$0.223	84.5%	2.1%
$0.0431	$0.350	77.0%	0.9%
$0.0462	$0.329	73.8%	0.1%
$0.0498	$0.279	65.8%	3.6%
$0.0523	$0.359	72.7%	3.4%
$0.0827	$0.279	17.6%	0.0%
$0.0409	$0.280	72.2%	0.4%

Effective in $0.0130Effective out $0.280
Token share
89.5%
Effective in $0.0194Effective out $0.223
Token share
2.1%
Effective in $0.0431Effective out $0.350
Token share
0.9%
Effective in $0.0462Effective out $0.329
Token share
0.1%
Effective in $0.0498Effective out $0.279
Token share
3.6%
Effective in $0.0523Effective out $0.359
Token share
3.4%
Effective in $0.0827Effective out $0.279
Token share
0.0%
Effective in $0.0409Effective out $0.280
Token share
0.4%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

177tok/s

P50, best across providers

Latency

0.51s

P50, best provider

Latency / throughput

AutoExacto benchmarks

GPQA DiamondTAU-BenchVenice85.8%--Novita78.3%78.2%Xiaomi78.4%76.2%GMICloud78.5%75.9%Io Net76.2%76.3%

+6 more providers

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)

100.00%

Availability (3d)

99.54%

Availability over the last 3 days

Last 72 hours

Availability 99.54%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.49%

Without Routing

87.21%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Xiaomi: MiMo-V2.6-Flash (Artificial Analysis and Vals AI)
Source	Benchmark	Score
Artificial Analysis	MiMo-V2.6-Flash Intelligence Index	26.7
Artificial Analysis	MiMo-V2.6-Flash HLE	35.1%
Artificial Analysis	MiMo-V2.6-Flash AA-LCR	74.3%
Artificial Analysis	MiMo-V2.6-Flash GDPval-AA	55.7%
Artificial Analysis	MiMo-V2.6-Flash CritPt	12.0%
Artificial Analysis	MiMo-V2.6-Flash SciCode	51.3%
Artificial Analysis	MiMo-V2.6-Flash Terminal-Bench 4.0	22.7%
Artificial Analysis	MiMo-V2.6-Flash AA-Omniscience Accuracy	27.2%
Artificial Analysis	MiMo-V2.6-Flash AA-Omniscience Non-Hallucination Rate	47.8%
Vals AI	MiMo V2.6 Flash Access to Justice Benchmark	70.7%
Vals AI	MiMo V2.6 Flash BioMysteryBench	69.3%
Vals AI	MiMo V2.6 Flash Code Migration v1.1	39.6%
Vals AI	MiMo V2.6 Flash CyberBench v1.1	75.4%
Vals AI	MiMo V2.6 Flash EMB v1.1	68.9%
Vals AI	MiMo V2.6 Flash Finance Agent v2.1	63.3%
Vals AI	MiMo V2.6 Flash Vals Finance Index	57.1%
Vals AI	MiMo V2.6 Flash HLAB-Vals-1.1	13.3%
Vals AI	MiMo V2.6 Flash IOI	47.7%
Vals AI	MiMo V2.6 Flash Vals Legal Index	69.0%
Vals AI	MiMo V2.6 Flash Legal Research Bench	38.0%
Vals AI	MiMo V2.6 Flash MedCode	41.1%
Vals AI	MiMo V2.6 Flash MedScribe	85.3%
Vals AI	MiMo V2.6 Flash MysteryMechanism	21.6%
Vals AI	MiMo V2.6 Flash ProgramBench	0.5%
Vals AI	MiMo V2.6 Flash ProofBench v1.1	63.0%
Vals AI	MiMo V2.6 Flash Public Benefits Bench v1.1	67.6%
Vals AI	MiMo V2.6 Flash SAGE	83.3%
Vals AI	MiMo V2.6 Flash Vals Science Index	31.7%
Vals AI	MiMo V2.6 Flash SRE Bench	4.2%
Vals AI	MiMo V2.6 Flash Tax Agent Bench	28.5%
Vals AI	MiMo V2.6 Flash Terminal-Bench 4.0	24.2%
Vals AI	MiMo V2.6 Flash Terminal-Bench-Science	4.3%
Vals AI	MiMo V2.6 Flash Vals Index	54.8%
Vals AI	MiMo V2.6 Flash Vibe Code Bench v1.1	79.0%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Freebuff

Powerful coding models, funded by ads

1.56Ttokens

Cline

Cline is an open-source AI coding agent that lives inside your IDE, autonomously exploring your codebase, editing files, running terminal commands, and using browser automation.

1.22Ttokens

Hello Minds, powered by Ethoswarm

new

709Btokens

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

465Btokens

ISEKAI ZERO

AI adventures. Travel with your favorite characters

373Btokens

Rest of OpenRouter

467B

Top 5 Apps

38.8B

Not currently available through EU in-region routing. In-region routing docs

More models from Xiaomi

MiMo-V2.6-Pro-UltraSpeed

MiMo-V2.6-Pro-UltraSpeed is the fast speed edition of Xiaomi's flagship foundation model, MiMo-V2.6-Pro. Built from the same 1T MiMo-V2.6-Pro checkpoint, it matches the original model in quality while delivering roughly 10x the output speed. The model features a 1M-token context window and native multimodal capabilities. Optimized for agentic workflows, it delivers top-tier performance across coding, visual, general, and research scenarios, excelling at complex, long-horizon tasks with robust generalization across a diverse range of agent harnesses.

Text1.0M context$4.35 / $8.70

MiMo-V2.6-Pro

MiMo-V2.6-Pro is the flagship foundation model developed by Xiaomi. Built at a scale of over 1T parameters, it is designed to push the ceiling of capability for the most demanding workloads. The model features a 1M-token context window and native multimodal capabilities. Optimized for agentic workflows, it delivers top-tier performance across coding, visual, general, and research scenarios, excelling at complex, long-horizon tasks with robust generalization across a diverse range of agent harnesses.

Text1.1M context$0.3263 / $0.6525

MiMo-V2.5-Pro

MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as ClawEval, GDPVal, and SWE-bench Pro. It can independently and autonomously complete professional tasks that would take human experts days or weeks, involving more than a thousand tool calls. Its context length of up to 1M makes it well suited for integration with a wide range of agent frameworks.

Text1.0M context$0.435 / $0.87

MiMo-V2.5

MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding tasks. Its 1M context window supports complete documents, extended conversations, and complex task contexts in a single pass, making it ideal for integration with agent frameworks where strong reasoning, rich perception, and cost efficiency all matter.

Text1M context$0.40 / $2

MiMo-V2-Omni

MiMo-V2-Omni is a frontier omni-modal model that natively processes image, video, and audio inputs within a unified architecture. It combines strong multimodal perception with agentic capability - visual grounding, multi-step planning, tool use, and code execution - making it well-suited for complex real-world tasks that span modalities, 256K context window.

Text262K context

MiMo-V2-Pro

MiMo-V2-Pro is Xiaomi's flagship foundation model, featuring over 1T total parameters and a 1M context length, deeply optimized for agentic scenarios. It is highly adaptable to general agent frameworks like OpenClaw. It ranks among the global top tier in the standard PinchBench and ClawBench benchmarks, with perceived performance approaching that of Opus 4.6. MiMo-V2-Pro is designed to serve as the brain of agent systems, orchestrating complex workflows, driving production engineering tasks, and delivering results reliably.

Text1.0M context

MiMo-V2-Flash

MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi. It is a Mixture-of-Experts model with 309B total parameters and 15B active parameters, adopting hybrid attention architecture. MiMo-V2-Flash supports a hybrid-thinking toggle and a 256K context window, and excels at reasoning, coding, and agent scenarios. On SWE-bench Verified and SWE-bench Multilingual, MiMo-V2-Flash ranks as the top #1 open-source model globally, delivering performance comparable to Claude Sonnet 4.5 while costing only about 3.5% as much.

Users can control the reasoning behaviour with the reasoning enabled boolean. Learn more in our docs.

Text262K context

Xiaomi: MiMo-V2.6-Flash

Providers