Towards AITowards AIToneBench

Model Comparison

Claude Opus 4.8 vs MiMo-V2.6-Pro UltraSpeed

Overlapping ranges: the best configurations, Claude Opus 4.8 (max effort) and MiMo-V2.6-Pro UltraSpeed (thinking on), are 13 Elo apart, and their 95% ranges overlap. Among 87 models, each ranked by its best configuration, Claude Opus 4.8 is #19 and MiMo-V2.6-Pro UltraSpeed is #21. Cost per script: $1.23 for Claude Opus 4.8 (max effort) and $0.177 for MiMo-V2.6-Pro UltraSpeed (thinking on).

Last updated , from board data generated at 17:52 UTC.

Claude Opus 4.8
1926 Writing Elo of Claude Opus 4.8 (max effort), its best configuration#19 of 87 models · 80.7/100 Writing Score · $1.23 per script2 configurations on the board
MiMo-V2.6-Pro UltraSpeed
1913 Writing Elo of MiMo-V2.6-Pro UltraSpeed (thinking on), its best configuration#21 of 87 models · 80.6/100 Writing Score · $0.177 per script1 configuration on the board

Picks and Verdict

PickClaude Opus 4.8

Claude Opus 4.8 (max effort)

Pick Claude Opus 4.8 (max effort) when length and YouTube structure are what you fix most, and $1.23 a script is acceptable.

PickMiMo-V2.6-Pro UltraSpeed

MiMo-V2.6-Pro UltraSpeed (thinking on)

Pick MiMo-V2.6-Pro UltraSpeed (thinking on) as the everyday choice, because I'd take its price and speed over Opus at max effort and it costs $0.177 a script.

Claude Opus 4.8 and MiMo-V2.6-Pro UltraSpeed land 13 Writing Elo apart, and I still wouldn't pick Opus here. Its max effort setting sits at 1926 and MiMo-V2.6-Pro UltraSpeed (thinking on) at 1913, with overlapping 95% ranges, though that alone does not test their difference. That's the #19 model of 87 against the #21 model, each ranked by its best setting. Claude Opus 4.8 (max effort) costs $1.23 a script, about seven times MiMo's $0.177, and takes nearly ten minutes where MiMo takes half of one. Dropping Opus to high fixes the price and loses the writing: Claude Opus 4.8 (high) costs a bit more than MiMo and falls 124 Elo behind it, with separate ranges. So the 137 Elo that max effort adds is what puts Opus back in MiMo's range. MiMo also beats Opus at max effort on Tone & Voice Match, craft and the Slop Score, and its drafts are a little steadier. Opus keeps the edge on structure: it's about five points better on Length Adherence and ahead on YouTube Best Practices. If hitting the exact length is the thing you fix most, that counts. For most scripts I'd run MiMo.

Disclosure: Claude Opus 4.8 comes from the same lab as one of the four judges, Claude Opus 5.5. The other three sit outside its lab, so it is never scored by its own lab alone (how the panel works). The judge analysis measures how much each judge favors its own lab.

How to Read the Writing Score, Writing Elo and 95% Ranges

Writing Score and Writing Elo. The Writing Score (0 to 100) is the weighted average of the nine metric scores: seven from the judges, Length Adherence from code and the Slop Score half from each. The Writing Elo is a rating from pairwise comparisons: on each script, each configuration's average across its drafts is compared with every other configuration's, and gaps within the draft-to-draft variation count as draws. Elo rewards winning often, so the two can put configurations in a different order.

Elo ranges. Each 95% range shows how far a configuration's Elo could move if the drafts had come out slightly differently. It comes from resampling every configuration's recorded draft scores on each reference script and ranking again. Comparing two separate ranges does not test the Elo gap between them: ranges far apart are strong evidence for the order, and overlapping ranges do not mean a tie. The reference scripts stay fixed, so the ranges say nothing about new scripts. The methodology page has the full description.

All Thinking Levels of Both Models

Every ranked configuration of Claude Opus 4.8 and MiMo-V2.6-Pro UltraSpeed, scored on the same 10 scripts by the same four judges.

How to Read the Chart
  • Each row is one configuration: the model at one thinking level. The dot is its Writing Elo, a rating from pairwise comparisons of the scores on each script, and the bar is its 95% range: how far that Elo could move if the drafts had come out slightly differently.
  • Each bar describes one configuration. Overlap alone does not establish a tie or decide their order. Comparing these ranges does not directly test the difference between a pair.
  • The table under the chart lists the same configurations in the same order and adds the Writing Score (the 0 to 100 consensus of the judges and code), cost, time and configuration rank (the place among all configurations on the board).
Claude Opus 4.8MiMo-V2.6-Pro UltraSpeed
ConfigurationWriting Elo95% RangeWriting ScoreCost / ScriptTime / ScriptReasoning TokensConfiguration Rank
Claude Opus 4.8 (max effort)19261889 to 195880.7$1.239.8 minN/A#39
MiMo-V2.6-Pro UltraSpeed (thinking on)19131862 to 195280.6$0.1770.5 min10,745#43
Claude Opus 4.8 (high)17891749 to 182979.3$0.1931.4 minN/A#59

Cost is one script at the API list price of the route used, uncached and without judging. Time is the average for one accepted script, retries included. N/A under Reasoning Tokens means the board has no separate reasoning count for that route.

Computed Findings Across Thinking Levels

Each finding is computed from the configurations above. The 95% ranges describe individual configurations; overlap alone establishes neither a tie nor an order for a pair.

Two Chosen Configurations Compared

Pick one configuration of each model to compare their metric scores, consistency across drafts and per-script scores. Each menu starts on the model's best configuration.

Scores by Metric

Claude Opus 4.8 (max effort)MiMo-V2.6-Pro UltraSpeed (thinking on)Solid bar: higher score in that row

Claude Opus 4.8 (max effort) scores higher on four of the nine metrics and MiMo-V2.6-Pro UltraSpeed (thinking on) on five. Weighted by each metric's share of the Writing Score, Length Adherence does the most to put Claude Opus 4.8 (max effort) ahead, with a 5.2-point lead on that metric. The weight beside each metric is its share of the Writing Score.

Tone & Voice Match19% weight
76.7
78.2
Writing Craft & Clarity13% weight
76.7
78.0
Substance, Accuracy & Value15% weight
82.0
82.7
Continuity & Emotion14% weight
80.9
80.0
YouTube Best Practices12% weight
80.4
78.6
Hook Strength10% weight
80.3
80.6
Length Adherence8% weight
94.6
89.4
Slop Score5% weight
82.1
83.8
Visual Cue Quality4% weight
79.7
79.3

Consistency Across Drafts

Claude Opus 4.8 (max effort)MiMo-V2.6-Pro UltraSpeed (thinking on)
Score spread across drafts (±)2.021.87
Lowest script score77.5 Script 478.5 Script 4
Highest script score83.1 Script 682.3 Script 7
Range across scripts5.63.8

Spread is how much the Writing Score moves between a configuration's drafts, and lower is steadier. Green marks the better number in each row.

Scores by Script

Writing Score on each reference script. Gap is Claude Opus 4.8 (max effort) minus MiMo-V2.6-Pro UltraSpeed (thinking on), and the higher score is in bold.

ScriptClaude Opus 4.8 (max effort)MiMo-V2.6-Pro UltraSpeed (thinking on)Gap
Script 1 opinion / warning explainer80.279.0+1.2
Script 2 news-analysis / skeptical explainer80.680.60.0
Script 3 personal roadmap / opinion79.380.9-1.6
Script 4 short explainer77.578.5-1.0
Script 5 news-analysis / opinion explainer80.381.2-0.9
Script 6 founder announcement / personal origin story83.181.2+1.9
Script 7 personal technical walkthrough / agentic workflow case study82.582.3+0.2
Script 8 product release announcement / personal observation82.581.3+1.2
Script 9 career guide / hiring analysis82.082.3-0.3
Script 10 engineering process walkthrough / presentation adaptation78.978.8+0.1

Compare With Another Model

Pick a second model. A pair with its own written match-up opens that page, and any other pair opens in the comparison tool.

Nearby Match-Ups by Best Writing Elo

Every configuration wrote five drafts for each of the same 10 briefs, from the same prompt. Generation was closed-book, with the prompt as the only source and no web search or other tools. Each judge scores every draft once: Claude Opus 5.5 at medium effort through the Claude Code subscription, GPT-6.1 Sol at medium reasoning through the Codex CLI, GLM 5.3 Flash through OpenRouter, and Jev 1.13, a typed judge from TypeSafe that answers each metric as a multiple-choice question. The channel's finished script is their reference for the voice, and it gets no score on the board itself.

← All comparisons