What are MiMo-V2.6-Flash's benchmark scores?

As of 9 Nov 2026, MiMo-V2.6-Flash scores 54.9 ±4.1 on the BenchLeader Index, strongest in maths (63) and weakest in knowledge (39). That rests on independent benchmark results only, and every score on this page links to its publisher.

How does MiMo-V2.6-Flash rank?

MiMo-V2.6-Flash is #216 of 599 ranked configurations on 9 Nov 2026. 272 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Fable 5.2 (max effort) at 69.4.

Is MiMo-V2.6-Flash better than Muse Spark 1.3?

As of 9 Nov 2026, Muse Spark 1.3 (max) leads on quality (61.7 vs 54.9 on the BenchLeader Index), but MiMo-V2.6-Flash costs 11× less, so it is the better buy unless you need the extra quality. The head-to-head page has the benchmark-by-benchmark detail.

When was MiMo-V2.6-Flash released?

Xiaomi released MiMo-V2.6-Flash on 21 Sep 2026. BenchLeader has results for it at 2 reasoning-effort settings.

Xiaomi Open weights: XiaomiMiMo/MiMo-V2.6-Flash-RLReasoning modelCompare with another model

MiMo-V2.6-Flash

Reasoning effort

Quality#107 of 335

54.9Index

Above the median

+4.9 on the median

Price#32 of 251

$0.175per 1M tokens

Cheapest quarter

0.2× the median

Speed#195 of 217

35tokens/s

Bottom quarter

0.5× the median

2 cheaper ranked models score higher: Step 3.5 Flash (56.4, $0.150), GPT-5 nano (55.9, $0.138).

As of 9 Nov 2026, MiMo-V2.6-Flash by Xiaomi scores 54.9 on the BenchLeader Index (599 ranked configurations), with its best results in maths (63) and weakest in knowledge (39). API pricing: $0.140 per 1M input tokens and $0.280 per 1M output tokens ($0.175 blended at 3:1), list price on 9 Nov 2026. It streams about 35 tokens/s (Artificial Analysis). Released 21 Sep 2026.

Key facts (as of 9 Nov 2026)

MiMo-V2.6-Flash API pricing: $0.140 per 1M input tokens and $0.280 per 1M output tokens ($0.175 blended at 3:1) (list price, 9 Nov 2026).
MiMo-V2.6-Flash streams about 35 tokens/s (Artificial Analysis, 9 Nov 2026).
MiMo-V2.6-Flash was released on 21 Sep 2026 (Xiaomi).

In detail. MiMo-V2.6-Flash is a Xiaomi open-weights reasoning model, released 21 Sep 2026. Its best configuration ranks #216 of 599 on the BenchLeader Index at 54.9 ±4.1, in the upper half. The ± is the point: 272 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 maths (63) and lowest in knowledge (39). At $0.175 per million tokens blended it is among the cheapest fifth of ranked models. Output speed of 35 tokens per second puts it in the slowest quarter, with a first answer in 45.7 s. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 8 Nov 2026.

Blended price: $0.175/M; $0.140 in · $0.280 out; via Artificial Analysis
Output speed: 35 tok/s; measured by Artificial Analysis
First answer: 46 s; first token 2.69 s
Context: 1.0M; via models.dev
Full answer: 127 s; median, reasoning included
Cost per task: $0.129; 62 min a task, AA Index
Tokens per task: 132k; 76% of it reasoning
Released: 21 Sep 2026

How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.

Overall index55
Reasoning48
Coding57
Agents & tools57
Maths63
Knowledge39
Human preference63
Multimodal57
Long context62

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	–	–	35 tok/s	46 s	$0.0001
not statedbest	54.9	#216	35 tok/s	46 s	$0.0001	Agents & tools 57 · Coding 57 · Human preference 63 · Knowledge 39 · Long context 62 · Maths 63 · Multimodal 57 · Reasoning 48

How its index has moved

26 Sep 2026 to 9 Nov 2026

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.

Metadata

Benchmark	not stated	Source	Trend
AA Openness Indexnot in index	52.8#26	Artificial Analysis

Reasoning

Benchmark	not stated	Source
Arena Hard Prompts	1473#80	Arena (LMArena)
Humanity's Last Exam (AA)not in index	35.1%#132	Artificial Analysis
HLE-Diamond (AA)	13.7%#104	Artificial Analysis
CritPtnot in index	12.0%#120	Artificial Analysis
CritPt v2 hard subset (AA)	29.5%#74	Artificial Analysis
MysteryMechanismnot in index	21.6%#23	Vals AI

Coding

Benchmark	no reasoning	not stated	Source
SciCodeeffort not stated	–	51.3%#106	SciCode
Arena Codingeffort not stated	–	1511#56	Arena (LMArena)
Arena WebDeveffort not stated	–	1636#22	Arena (LMArena)
SciCode (AA)not in indexeffort not stated	–	51.3%#110	Artificial Analysis
IOIeffort not stated	–	47.7%#28	Vals AI
Code Migrationnot in indexeffort not stated	–	40.9%#30	Vals AI
Code Migration v1.1not in indexeffort not stated	–	39.5%#21	Vals AI
ProgramBenchnot in indexeffort not stated	–	0.5%#19	Vals AI
Vibe Code Bench v1.1not in indexeffort not stated	–	79.0%#35	Vals AI
ALE-Benchnot in indexeffort not stated	–	1244.2#43	Epoch AI Benchmarking Hub
Terminal-Bench 4.0 (AA)not in indexeffort not stated	–	22.7%#79	Artificial Analysis
GameBench 2 by reasoning settingnot in index	39.4#123	–	DuelLab GameBench 2
FrontierCode 1.1 (AA)not in indexeffort not stated	–	42.1%#32	Artificial Analysis
FrontierCode 2.0 (AA)effort not stated	–	21.6%#60	Artificial Analysis

Agents & tools

Benchmark	not stated	Source
Arena Agent	0.1#29	Arena (LMArena)
Terminal-Bench 2.1 (Vals)	76.4%#16	Vals AI
GDPval-AA v2.1, normalised scorenot in index	55.4%#23	Artificial Analysis
GDPval-AA v2.1not in index	1614#26	Artificial Analysis
CyberBenchnot in index	75.4%#7	Vals AI
Finance Agent v2not in index	56.3%#22	Vals AI
Finance Agent v2.1not in index	63.3%#16	Vals AI
Harvey's Legal Agent Benchmark v1.1not in index	13.3%#15	Vals AI
Harvey's Legal Agent Benchmark v1not in index	11.3%#15	Vals AI
Legal Research Benchnot in index	38.0%#43	Vals AI
SREBenchnot in index	4.2%#19	Vals AI
Tax Agent Benchnot in index	28.5%#37	Vals AI
Terminal-Bench 4.0 (Vals)not in index	24.2%#32	Vals AI
Terminal-Bench Sciencenot in index	4.3%#27	Vals AI
AutomationBenchnot in index	64.0%#35	Artificial Analysis
GDP.pdf (AA)not in index	9.0%#163	Artificial Analysis
AA-Briefcase v1.1not in index	1487#46	Artificial Analysis
Terminal-Bench-Science 0.1 (AA)not in index	4.3%#81	Artificial Analysis

Maths

Benchmark	not stated	Source	Trend
ProofBench	63.0%#21	Vals AI
Arena Maths	1468#58	Arena (LMArena)

Knowledge

Benchmark	not stated	Source
AA-Omniscience v1.1	-10.8#99	Artificial Analysis
BioMysteryBenchnot in index	69.3%#16	Vals AI
Excel Modeling Benchmarknot in index	65.5%#24	Vals AI
Excel Modeling Benchmark v1.1not in index	68.9%#25	Vals AI
MedCodenot in index	41.1%#58	Vals AI
MedScribenot in index	85.3%#30	Vals AI
Public Benefits Bench v1.1not in index	67.6%#14	Vals AI
AA-Omniscience v1.1: accuracynot in index	27.2%#97	Artificial Analysis
AA-Omniscience v1.1: non-hallucinationnot in index	47.8%#75	Artificial Analysis

Instruction following

Benchmark	not stated	Source	Trend
Arena Instruction Followingnot in index	1437#82	Arena (LMArena)

Human preference

Benchmark	not stated	Source
Arena Text	1448#83	Arena (LMArena)
Arena Creative Writingnot in index	1392#121	Arena (LMArena)
Arena Multi-turnnot in index	1447#96	Arena (LMArena)
Arena Longer Queriesnot in index	1458#78	Arena (LMArena)

Multimodal

Benchmark	not stated	Source
Arena Vision	1259#56	Arena (LMArena)
MMMU-Pro	73.1%#139	Artificial Analysis
SAGEnot in index	83.3%#22	Vals AI
Code2Video-AAnot in index	1052#10	Artificial Analysis

Long context

Benchmark	not stated	Source	Trend
AA-LCR v1.1	74.3%#177	Artificial Analysis
MLCRnot in index	11.7%#105	Artificial Analysis

Composite

Benchmark	not stated	Source
AA Intelligence Index v5.0not in index	26.7#71	Artificial Analysis
AA Intelligence Index v4.3.2not in index	37.9#85	Artificial Analysis
AA Progress Index v1not in index	311.8#90	Artificial Analysis
Vals Indexnot in index	54.8#23	Vals AI
Vals Finance Indexnot in index	57.1%#19	Vals AI
Vals Science Indexnot in index	31.7%#18	Vals AI
Vals Router Indexnot in index	54.8%#14	Vals AI

GameBench 2, best reasoning setting (None): 39.4. Reference only, not part of the BenchLeader Index.

DuelLab GameBench 2, release gb2-20261103-eaea31df94e6 (https://benchmarks.duellab.org/releases/gb2-20261103-eaea31df94e6/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

API pricing: where to run it

Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.

Provider	Speed	First token	Input $/M	Output $/M	Context	Quantisation
GMICloud	47 tok/s	2.88 s	$0.112	$0.224	1.1M	bf16
Makora	38 tok/s	0.86 s	$0.140	$0.280	1.0M	fp8
Cohere	37 tok/s	1.46 s	$0.140	$0.280	1.0M	fp8
Tenstorrent	31 tok/s	0.61 s	$0.165	$0.330	1.0M	fp8
DeepInfra	29 tok/s	3.53 s	$0.140	$0.280	1.0M	fp8
Xiaomi	28 tok/s	3.07 s	$0.140	$0.280	1.0M	fp8
Venice	27 tok/s	5.05 s	$0.175	$0.350	1M	fp8
Darkbloom	22 tok/s	1.94 s	$0.100	$0.280	1.0M	fp4

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 at $0.003 per 1M. Reasoning tokens are not modelled.

Workload	Tokens in / out	Cost	With caching	Time
Chat reply	400 / 300	$0.0001	$0.0001	54.2 s
Summarize a 30-page report	12,000 / 600	$0.0018	$0.0006	1.0 min
Code edit	6,000 / 1,500	$0.0013	$0.0006	1.5 min
Agentic coding session	60,000 / 4,000	$0.0095	$0.0033	2.6 min
Structured extraction	2,000 / 200	$0.0003	$0.0001	51.3 s

New this month

Compare MiMo-V2.6-Flash with…

Embed or cite

Show where MiMo-V2.6-Flash ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page. To see ranks like this on OpenRouter and Hugging Face model pages, get the Chrome extension.

[![MiMo-V2.6-Flash on BenchLeader: #106 overall](https://www.benchleader.com/badge/mimo-v2-6-flash.svg)](https://www.benchleader.com/models/mimo-v2-6-flash)

Cite this

The ranking changes daily, so a reference carries the data date. APA style:

BenchLeader. (2026, November 9). MiMo-V2.6-Flash on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 9, 2026, from https://www.benchleader.com/models/mimo-v2-6-flash

BibTeX:

@misc{benchleader_mimo_v2_6_flash_20261109,
  author       = {{BenchLeader}},
  title        = {{MiMo-V2.6-Flash on the BenchLeader Index (method v1.4)}},
  year         = {2026},
  month        = nov,
  howpublished = {\url{https://www.benchleader.com/models/mimo-v2-6-flash}},
  note         = {Data as of 2026-11-09}
}

Every badge style, the embeddable top-10 widget and how to cite the Index

Questions people ask

What are MiMo-V2.6-Flash's benchmark scores?: As of 9 Nov 2026, MiMo-V2.6-Flash scores 54.9 ±4.1 on the BenchLeader Index, strongest in maths (63) and weakest in knowledge (39). That rests on independent benchmark results only, and every score on this page links to its publisher.
How does MiMo-V2.6-Flash rank?: MiMo-V2.6-Flash is #216 of 599 ranked configurations on 9 Nov 2026. 272 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Fable 5.2 (max effort) at 69.4.
How much does MiMo-V2.6-Flash cost per 1M tokens?: As of 9 Nov 2026, MiMo-V2.6-Flash costs $0.140 per 1M input tokens and $0.280 per 1M output tokens, $0.175 blended (three input tokens per output token).
Is MiMo-V2.6-Flash better than Muse Spark 1.3?: As of 9 Nov 2026, Muse Spark 1.3 (max) leads on quality (61.7 vs 54.9 on the BenchLeader Index), but MiMo-V2.6-Flash costs 11× less, so it is the better buy unless you need the extra quality. The head-to-head page has the benchmark-by-benchmark detail.
When was MiMo-V2.6-Flash released?: Xiaomi released MiMo-V2.6-Flash on 21 Sep 2026. BenchLeader has results for it at 2 reasoning-effort settings.

Data as of 9 Nov 2026. Compare these configurations.

Cite as: BenchLeader, “MiMo-V2.6-Flash: benchmarks, pricing, speed and rank”, data as of 9 Nov 2026, https://www.benchleader.com/models/mimo-v2-6-flash.