Xiaomi model product
MiMo V2 Flash is Xiaomi's Mixture-of-Experts foundation language model with 309B total parameters, 15B active parameters, and a 256k context window.
Updated Sep 8, 2026. Default version: MiMo-V2-Flash
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for MiMo-V2-Flash.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkArena-Hard v2 | Score86.20% | Rank01 | Participants19 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTau-bench | Score80.30% | Rank03 | Participants6 | Percentile60.00% | EvidenceC | Evaluated |
| BenchmarkMRCR | Score45.70% | Rank06 | Participants7 | Percentile16.67% | EvidenceC | Evaluated |
| BenchmarkLongBench v2 | Score60.60% | Rank09 | Participants17 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkSWE-bench Multilingual | Score71.70% | Rank19 | Participants43 | Percentile57.14% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench | Score30.50% | Rank19 | Participants25 | Percentile25.00% | EvidenceC | Evaluated |
| BenchmarkHMMT 2025 | Score84.40% | Rank22 | Participants33 | Percentile34.38% | EvidenceC | Evaluated |
| BenchmarkAIME 2025 | Score94.10% | Rank24 | Participants119 | Percentile80.51% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench v6 | Score80.60% | Rank24 | Participants62 | Percentile62.30% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score84.90% | Rank25 | Participants138 | Percentile82.48% | EvidenceC | Evaluated |
| BenchmarkBrowseComp | Score58.30% | Rank39 | Participants63 | Percentile38.71% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score73.40% | Rank46 | Participants113 | Percentile59.82% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.0 | Score38.50% | Rank47 | Participants51 | Percentile8.00% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam | Score22.10% | Rank63 | Participants103 | Percentile39.22% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score83.70% | Rank72 | Participants247 | Percentile71.14% | EvidenceC | Evaluated |
| BenchmarkLM Arena Webdev | Score1,330.71 rating | Rank80 | Participants97 | Percentile17.71% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,421.89 rating | Rank85 | Participants121 | Percentile30.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,411.21 rating | Rank93 | Participants210 | Percentile55.98% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,392.15 rating | Rank116 | Participants210 | Percentile44.98% | EvidenceA | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext factuality | Categorygerman | Rank36 | Rating / score1,389.04 | Votes403 | ObservationsN/A | Result date |
| Arenatext factuality | Categorypolish | Rank36 | Rating / score1,403.39 | Votes750 | ObservationsN/A | Result date |
| Arenatext factuality | Categorychinese | Rank50 | Rating / score1,478.61 | Votes1,781 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryspanish | Rank50 | Rating / score1,424.81 | Votes974 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryfrench | Rank59 | Rating / score1,431.41 | Votes872 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry legal and government | Rank65 | Rating / score1,436.13 | Votes2,101 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry medicine and healthcare | Rank67 | Rating / score1,454.47 | Votes1,931 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank68 | Rating / score1,465.41 | Votes3,022 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry mathematical | Rank72 | Rating / score1,403.51 | Votes1,551 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryexpert | Rank73 | Rating / score1,442.88 | Votes2,393 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryhard prompts english | Rank73 | Rating / score1,462.67 | Votes8,645 | ObservationsN/A | Result date |
| Arenatext factuality | Categorycoding | Rank75 | Rating / score1,477.45 | Votes8,504 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymath | Rank75 | Rating / score1,402.99 | Votes1,763 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-react | Rank75 | Rating / score1,309.57 | Votes2,270 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank76 | Rating / score1,420.80 | Votes1,503 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-html | Rank76 | Rating / score1,353.83 | Votes3,130 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry life and physical and social science | Rank77 | Rating / score1,448.71 | Votes4,827 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank78 | Rating / score1,433.05 | Votes338 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank78 | Rating / score1,406.52 | Votes1,183 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry software and it services | Rank78 | Rating / score1,466.30 | Votes15,072 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank79 | Rating / score1,441.34 | Votes11,928 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank79 | Rating / score1,442.64 | Votes17,521 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank80 | Rating / score1,426.89 | Votes3,727 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank80 | Rating / score1,435.00 | Votes11,804 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryenglish | Rank80 | Rating / score1,440.35 | Votes18,117 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry business and management and financial operations | Rank80 | Rating / score1,426.53 | Votes5,987 | ObservationsN/A | Result date |
| Arenawebdev | Categoryoverall | Rank80 | Rating / score1,330.71 | Votes5,403 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev | Rank80 | Rating / score1,330.71 | Votes5,403 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank81 | Rating / score1,430.01 | Votes19,216 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry entertainment and sports and media | Rank81 | Rating / score1,389.00 | Votes5,894 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderQiniu | Provider model IDmimo-v2-flash | Regionglobal | Input / 1M$0.10 | Output / 1M$0.30 | Context256K | Updated |
| ProviderQiniu | Provider model IDxiaomi/mimo-v2-flash | Regionglobal | Input / 1M$0.10 | Output / 1M$0.30 | Context256K | Updated |
| ProviderMeganova | Provider model IDXiaomiMiMo/MiMo-V2-Flash | Regionglobal | Input / 1M$0.10 | Output / 1M$0.30 | Context262.1K | Updated |
| ProviderHugging Face | Provider model IDXiaomiMiMo/MiMo-V2-Flash | Regionglobal | Input / 1M$0.10 | Output / 1M$0.30 | Context262.1K | Updated |
| ProviderZenMux | Provider model IDxiaomi/mimo-v2-flash | Regionglobal | Input / 1M$0.10 | Output / 1M$0.30 | Context262.1K | Updated |
| ProviderXiaomi | Provider model IDmimo-v2-flash | Regionglobal | Input / 1M$0.14 | Output / 1M$0.28 | Context262.1K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for MiMo-V2-Flash. Runtime does not affect the capability score.
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionMiMo-V2-Flash | Released | LLMBoard50.97 | Parameters309B | Context256K | Max output16.4K | Open weightsYes | LicenseMIT |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about MiMo V2 Flash and its available data.
MiMo V2 Flash is a foundation language model from Xiaomi for reasoning, coding, and agentic scenarios. It uses a hybrid attention architecture with sliding-window and full attention in a 5:1 ratio, a 128-token window, and delivers 150 tokens/sec inference.
Data as of 2026-09-08.
Common questions about MiMo V2 Flash.
MiMo V2 Flash's default version was released on Dec 16, 2025.
MiMo V2 Flash's official API price is $0.14 per million input tokens and $0.28 per million output tokens via Xiaomi. The lowest tracked third-party offer starts at $0.10 input and $0.30 output via Qiniu.
MiMo V2 Flash was created by Xiaomi.
The default version has a 256K token context window.
Yes. The default version is marked as open weight under MIT.
6 provider offerings are linked to the default version.