ai MiMo-V2.6-Flash
Other · open weights · writing benchmark
The short version
MiMo-V2.6-Flash sits at #36 of 167 on ToneBench, in the middle of the pack, with a writing Elo of 2113 and an overall score of 87.3 out of 100. We measured it by having it write all 10 of our real YouTube scripts, five times each, then scoring every draft blind against our own finished versions. Here is how it shook out.
Within Other it ranks 3rd of 9. MiMo-V2.6-Pro UltraSpeed is the family's top writer here, about 84 Elo ahead, so this config trades some quality for whatever it saves you in cost or speed.
Among open-weights models, it comes in 5th of 53. That is a genuinely strong showing for an open-weights model on a voice-and-tone task, which is usually where the closed frontier still pulls ahead.
One thing that stands out: it is unusually consistent. Across its 50 drafts the overall score only moves ± 1.3 points run to run, versus a board median of ± 2.7. What you get on the first try is close to what you get every try.
What it does best is Hook: it ranks 31st on that dimension at 88.3, well above the board average. Its softer spot is Substance & Value (86.5, 53rd), which is the thing to watch if that metric matters most for your use.
It held steady across all 10 scripts, with its best run on the career guide / hiring analysis (88.1). Consistency like that is underrated, honestly, since it means fewer surprises when you actually put it to work.
On cost, it runs about $0.004/task. For that money it beats 100 models that cost noticeably more.
Skill profile
The shape is the story: the further a corner reaches, the stronger that dimension. Biggest edge over the board: Length Discipline (+15.9); smallest edge: Hook (+6.7).
Per-metric scores
Nine writing dimensions, each scored 0–100 and blended by the editorial weight shown. Rank is against all 167 current-ranked models. The thick bar is this model; the thin lines above and below are the current field's best model (darker) and average (lighter) on the same scale. The small ± number is its run-to-run variation.
Per-article scores
The same 10 real scripts every current-ranked model writes, scored individually. Different formats stress different skills.
Consistency: run-to-run variability
Every cell on this page is the mean of 5 independent runs per script; the ± numbers are the run-to-run standard deviation. Overall, MiMo-V2.6-Flash varies by ± 1.3 points between runs versus a board median of ± 2.7, so it is noticeably steadier than the typical model here.
| Metric | This model | Board median | Verdict |
|---|---|---|---|
| Tone & Voice ? | 87.6 ± 1.2 | ± 1.7 | typical |
| Writing Craft ? | 88.2 ± 0.9 | ± 1.5 | typical |
| Substance & Value ? | 86.5 ± 1.4 | ± 2.0 | typical |
| Flow & Emotion ? | 86.5 ± 1.4 | ± 1.8 | typical |
| YouTube Structure ? | 86.4 ± 2.6 | ± 2.9 | typical |
| Hook ? | 88.3 ± 1.3 | ± 2.0 | typical |
| Length Discipline ? | 86.7 ± 5.9 | ± 8.6 | typical |
| Anti-Slop ? | 91.3 ± 1.3 | ± 2.1 | typical |
| Visual Cues ? | 86.8 ± 2.2 | ± 3.2 | typical |
Measured, not modeled
Numbers we log on every run and rarely talk about. None of these affect the writing scores; cost and latency are informational.
What each judge scored it
The published overall of 87.3 is the consensus of three family-disjoint judges scoring the same 50 stored drafts blind with the identical rubric. The highest and lowest judge differ by 4.1 points on its overall. The three judges essentially agree on this model. How the panel works: methodology.
How we ran MiMo-V2.6-Flash
- Accessed viaOpenRouter · model page ↗
- Exact model / route id
xiaomi/mimo-v2.6-flash(provider default reasoning) - Generations5 independent runs × 10 scripts = 50 drafts, stored immutably after generation
- Run dates2026-09-25
- Judges3-judge panel: Claude Opus 5, high effort (Anthropic) + DeepSeek V4.1 Flash (OpenRouter) + GPT-5.6 Sol, medium reasoning (Codex CLI): every draft scored independently by all three with the identical rubric, blind to model identity, evidence quotes required before each metric score; published score = per-metric consensus mean, verified with a Bradley-Terry refit (methodology)
- Samplingprovider-default sampling, no fine-tuning, no system-prompt changes; every current-ranked model gets the identical brief, style contract and research packet
- What stays privatethe exact rubric and any non-public reference material; source-publication timing is disclosed for exceptions (why)
Frequently asked questions
How good is MiMo-V2.6-Flash at writing?
On ToneBench it ranks #36 of 167 with a writing Elo of 2113 and an overall score of 87.3/100. That score comes from writing our 10 real YouTube scripts five times each and scoring every draft blind against our own finished versions across nine writing dimensions.
Is MiMo-V2.6-Flash the best Other model for writing?
Not quite. Within Other it ranks 3rd of 9; MiMo-V2.6-Pro UltraSpeed is the family's best writer here.
Is MiMo-V2.6-Flash good value for the money?
It costs about $0.004/task. Nothing meaningfully cheaper outscores it, which puts it on the value side of the board.
What are MiMo-V2.6-Flash's strengths and weaknesses?
Its strongest dimension is Hook (31st on the board, 88.3). Its weakest is Substance & Value (53rd on the board, 86.5). The full nine-metric breakdown is on this page.
Does MiMo-V2.6-Flash write some formats better than others?
It was fairly even across all 10 scripts, topping out on the career guide / hiring analysis (88.1).
How consistent is MiMo-V2.6-Flash between runs?
We run every script 5 times. MiMo-V2.6-Flash's overall score varies by about ±1.3 points between runs, versus a board median of ±2.7. That makes it one of the steadier models we test. The full per-metric variability table is on this page.
How was MiMo-V2.6-Flash evaluated?
Via OpenRouter using the exact model/route id xiaomi/mimo-v2.6-flash, run on 2026-09-25. 5 runs per script, provider-default sampling, no fine-tuning; every draft scored blind with a fixed rubric by a three-family judge panel. Full details in the 'How we ran it' section and the methodology.
Is MiMo-V2.6-Flash open source?
Yes, it is an open-weights model. Among open-weights models it ranks 5th of 53 for writing.