Compare models

Context, limits, capabilities, price and use cases side by side — up to 4 models at a time. The URL holds your selection, so you can share it.

MiMo V2.6 ProMiMo V2.6 Flash Add a modelChat with these 2

AttributeHide identical rows	MiMo V2.6 Proxiaomi/mimo-v2.6-pro	MiMo V2.6 Flashxiaomi/mimo-v2.6-flash
Specs
Context window	1M	1M
Max output	131K	131K
Input modalities	TextImageAudioVideo	TextImageAudioVideo
Output modalities	Text	Text
Capabilities
Streaming	Supported	Supported
Tool calling	Supported	Supported
Structured output (JSON)	Supported	Supported
Image input	Supported	Supported
Thinking mode	Supported	Supported
Price / 1M tokens
Input	$ 0.47	$ 0.16
Output	$ 0.94	$ 0.32
Cache read	$ 0.0050	$ 0.0040
Cache write	$ 0.47	$ 0.16
What it is good at
Best for	Long-horizon coding agents Cybersecurity and scientific research Omni-modal agents	High-frequency professional workloads Coding agents Omni-modal understanding
Published benchmarks
Terminal-Bench 2.1pass_rate	89.9%(setup not fully disclosed)	87.6%(setup not fully disclosed)
Terminal-Bench 4.0pass_rate	34.9%(setup not fully disclosed)	28.8%(setup not fully disclosed)
AutomationBench v1.0.6score	53.1%(setup not fully disclosed)	52.3%(setup not fully disclosed)
DeepSWE 1.1resolved_rate	71.9%(setup not fully disclosed)	67.9%(setup not fully disclosed)
ProgramBenchscore	26.5%(setup not fully disclosed)	26%(setup not fully disclosed)
OSWorld-Verifiedsuccess_rate	82%(setup not fully disclosed)	80.8%(setup not fully disclosed)
CyberGymscore	94%(setup not fully disclosed)	95.1%(setup not fully disclosed)
ExploitBenchscore	47.9%(setup not fully disclosed)	25.3%(setup not fully disclosed)
Toolathlon Verifiedpass_rate	76.9%(setup not fully disclosed)	73.6%(setup not fully disclosed)