Lab Comparison
Last updated , from board data generated at 17:52 UTC.
All 3 ranked Xiaomi configurations on ToneBench, side by side: Writing Elo, Writing Score, the nine metrics, and cost and time per script. A configuration is one model at one thinking level.
Ranks in these notes are configuration ranks: each setting's place among all 151 configurations. The Highest Writing Elo card above gives the model rank.
Picks and Notes From Louis
I'd start with MiMo-V2.6-Flash (thinking on): $0.0057 per script, open weights, and #44 on the board. It's hard to argue with that price.
MiMo-V2.6-Pro UltraSpeed (thinking on) tops the family on paper, and what its $0.177 buys is speed: 0.5 minutes per script against 3.5 for the Flash. That's 31.0 times the money, so pay for UltraSpeed only when waiting is the actual problem.
The one I'd skip is MiMo-V2.6-Pro (thinking on). It costs more than the Flash, takes 6.2 minutes per script, and doesn't rank higher.
You probably know Xiaomi for phones, but the MiMo models write good scripts. All 3 configurations sit within 42 Elo of each other, from #43 to #51, and their Elo ranges all overlap. That overlap alone doesn't say which of them writes better. The choice comes down to price and speed. Only MiMo-V2.6-Flash (thinking on) ships as open weights, which matters if you'd rather run it yourself.
Every row wrote the same 10 scripts for the same four blind judges, so these Writing Elo ratings sit on the leaderboard's scale. The rank column is the configuration rank, each setting's place among all 151 configurations on the board. Each row is a different model. The ± after each Elo is half the width of its 95% range, and scores run from 0 to 100. Cost is one full script at the API list price of the route used. Each name opens its scorecard, and the head-to-head comparisons cover models from different labs.
With the metric scores shown, the table scrolls sideways.
| Configuration Rank | Configuration | Writing Elo | Writing Score | Cost / Script | Time / Script | Tone & Voice Match | Writing Craft & Clarity | Substance, Accuracy & Value | Hook Strength | YouTube Best Practices | Continuity & Emotion | Slop Score | Length Adherence | Visual Cue Quality |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| #43 | MiMo-V2.6-Pro UltraSpeed (thinking on) | 1913 ±45 | 80.6 | $0.177 | 0.5 min | 78.2 | 78.0 | 82.7 | 80.6 | 78.6 | 80.0 | 83.8 | 89.4 | 79.3 |
| #44 | MiMo-V2.6-Flash (thinking on) · open weights | 1902 ±50 | 80.4 | $0.0057 | 3.5 min | 77.3 | 78.4 | 81.0 | 82.7 | 79.2 | 80.2 | 84.1 | 87.6 | 79.5 |
| #51 | MiMo-V2.6-Pro (thinking on) | 1871 ±43 | 80.2 | $0.017 | 6.2 min | 78.0 | 77.8 | 80.9 | 79.7 | 78.8 | 79.8 | 83.3 | 90.9 | 78.3 |
Each comparison lists every thinking level of both models, with Writing Elo, cost and time side by side. The line under each pair gives the Writing Elo gap between the two best configurations. A pair outside the written match-ups opens in the comparison tool.