XiaomiMiMo-V2-Flash
MiMo-V2-Flash is a Xiaomi open-weights reasoning model, released 16 Dec 2025. Its best configuration (thinking reasoning effort) ranks #318 of 785 on the BenchLeader Index at 52.4 ±5.1, in the upper half. The ± is the point: 313 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (64) and lowest in coding (40). At $0.150 per million tokens blended it is among the cheapest fifth of ranked models. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 22 Oct 2026.
- Blended price
- $0.150/M
- $0.100 in · $0.300 out
- Output speed
- –
- First answer
- –
- Context
- 256k
- Full answer
- –
- Cost per run
- –
- Released
- 16 Dec 2025
- Overall index52
- Reasoning51
- Coding40
- Agents & tools58
- Maths54
- Knowledge41
- Instruction following64
- Human preference56
- Long context60
- Composite53
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | 48.2 | #434 | – | – | – | Agents & tools 56 · Coding 44 · Composite 47 · Human preference 56 · Instruction following 43 · Knowledge 40 · Long context 42 · Maths 55 · Reasoning 46 |
| thinkingbest | 52.4 | #318 | – | – | $0.0001 | Agents & tools 58 · Coding 40 · Composite 53 · Human preference 56 · Instruction following 64 · Knowledge 41 · Long context 60 · Maths 54 · Reasoning 51 |
| not stated | 51.6 | #339 | – | – | – | Agents & tools 61 · Coding 21 · Composite 55 · Instruction following 70 · Knowledge 53 · Long context 61 · Reasoning 46 |
How its index has moved
9 Sept 2026 to 21 Oct 2026- thinking
- not stated
- no reasoning
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| Arena Hard Prompts | 1413#160 | 1411#163 | – | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 65.6%#341 | 84.7%#135 | 83.5%#156 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 8.6%#348 | 22.9%#197 | 22.1%#202 | Artificial Analysis | |
| GPQA Diamond (Vals)not in indexeffort not stated | – | – | 59.3%#105 | Vals AI | |
| CritPt | 0.0%#315 | 4.3%#166 | 2.9%#182 | Artificial Analysis |
Coding
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| SciCodeeffort not stated | – | – | 25.9%#206 | SciCode | |
| Arena Coding | 1447#148 | 1432#169 | – | Arena (LMArena) | |
| Arena WebDev | 1331#114 | 1293#119 | – | Arena (LMArena) | |
| ALE-Benchnot in indexeffort not stated | – | – | 738.0#83 | Epoch AI Benchmarking Hub | |
| Terminal-Bench 2.1 (AA)not in index | 61.8%#92 | – | – | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| Terminal-Bench Hard | 25.8%#135 | 28.0%#124 | 31.1%#106 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 83.9%#105 | 95.0%#26 | 93.3%#43 | Artificial Analysis |
Maths
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| Arena Maths | 1376#175 | 1372#180 | Arena (LMArena) |
Knowledge
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| AA-Omniscience | -48.5#374 | -45.4#352 | -19.1#239 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 15.6%#407 | 24.8%#237 | 19.8%#310 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 24.0%#250 | 6.6%#500 | 51.6%#110 | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| IFBench | 39.9%#261 | 64.2%#112 | 71.8%#51 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1382#159 | 1375#167 | – | Arena (LMArena) |
Human preference
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| Arena Text | 1392#159 | 1386#169 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1353#164 | 1336#182 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1388#168 | 1375#175 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1403#155 | 1400#159 | Arena (LMArena) |
Long context
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| AA-LCR | 36.3%#372 | 70.7%#208 | 71.3%#203 | Artificial Analysis |
Composite
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| AA Intelligence Index v4.3.2 | 16.1#276 | 20.8#221 | 22.4#200 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0001 | – | – |
| Summarize a 30-page report | 12,000 / 600 | $0.0014 | – | – |
| Code edit | 6,000 / 1,500 | $0.0011 | – | – |
| Agentic coding session | 60,000 / 4,000 | $0.0072 | – | – |
| Structured extraction | 2,000 / 200 | $0.0003 | – | – |
See also
Questions people ask
- What are MiMo-V2-Flash's benchmark scores?
- As of 24 Oct 2026, MiMo-V2-Flash (thinking effort) scores 52.4 ±5.1 on the BenchLeader Index, strongest in instruction following (64) and weakest in coding (40). That rests on 11 independent benchmark results, and every score on this page links to its publisher.
- How does MiMo-V2-Flash rank?
- MiMo-V2-Flash (thinking effort) is #318 of 785 ranked configurations on 24 Oct 2026. 313 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does MiMo-V2-Flash cost per 1M tokens?
- As of 24 Oct 2026, MiMo-V2-Flash costs $0.100 per 1M input tokens and $0.300 per 1M output tokens, $0.150 blended (three input tokens per output token).
- When was MiMo-V2-Flash released?
- Xiaomi released MiMo-V2-Flash on 16 Dec 2025. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “MiMo-V2-Flash: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/mimo-v2-flash.