Towards AITowards AIToneBench

Model Comparison

Claude Opus 5.5 vs MiMo-V2.6-Pro

Claude Opus 5.5 (max effort) leads MiMo-V2.6-Pro (thinking on) by 843 Elo, and their 95% ranges sit 722 Elo apart. Even Claude Opus 5.5's lowest-rated configuration, Claude Opus 5.5 (low), ranks above MiMo-V2.6-Pro (thinking on). Among 84 models, each ranked by its best configuration, Claude Opus 5.5 is #1 and MiMo-V2.6-Pro is #24. Cost per script: $2.58 for Claude Opus 5.5 (max effort) and $0.017 for MiMo-V2.6-Pro (thinking on).

Last updated , from board data generated at 20:15 UTC.

Claude Opus 5.5
2763 Writing Elo of Claude Opus 5.5 (max effort), its best configuration#1 of 84 models · 85.9/100 Writing Score · $2.58 per script5 configurations on the board
MiMo-V2.6-Pro
1920 Writing Elo of MiMo-V2.6-Pro (thinking on), its best configuration#24 of 84 models · 80.2/100 Writing Score · $0.017 per script1 configuration on the board

Picks and Verdict

PickClaude Opus 5.5

Claude Opus 5.5 (xhigh)

Pick Claude Opus 5.5 (xhigh) for a script you'll publish, because it costs $1.00 and sits 301 Elo above Opus at low with separate ranges.

PickMiMo-V2.6-Pro

MiMo-V2.6-Pro (thinking on)

Pick MiMo-V2.6-Pro (thinking on) for bulk first drafts, where $0.017 a script lets you run many and keep the best.

Claude Opus 5.5 at its lowest setting finishes a script about four times faster than MiMo-V2.6-Pro and ranks 540 Elo higher, so MiMo's argument is price. It's a good one: $0.017 a script, where Claude Opus 5.5 (low) costs about 14 times as much. Claude Opus 5.5 is the #1 model of 84 and MiMo-V2.6-Pro is the #24 model: Opus at max effort holds 2763 Elo against 1920 for MiMo-V2.6-Pro (thinking on), and their 95% ranges sit far from each other. Against Opus at xhigh, MiMo trails on all nine metrics, by 7.5 points on substance. Inside the Opus ladder, low, medium and high have overlapping ranges, so on a tight budget I'd take low at $0.243 and skip the two in between. The setting that pays for itself is xhigh, 301 Elo above low with separate ranges. Max effort costs $2.58, about two and a half times xhigh, and their ranges overlap. I'd publish from Opus at xhigh and hand a batch of rough drafts to MiMo at under two cents.

Disclosure: Claude Opus 5.5 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 5.5 and MiMo-V2.6-Pro, 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 5.5MiMo-V2.6-Pro
ConfigurationWriting Elo95% RangeWriting ScoreCost / ScriptTime / ScriptReasoning TokensConfiguration Rank
Claude Opus 5.5 (max effort)27632687 to 283285.9$2.5817.5 minN/A#1
Claude Opus 5.5 (xhigh)27612688 to 283385.8$1.006.7 minN/A#2
Claude Opus 5.5 (high)25032450 to 254584.6$0.3282.1 minN/A#5
Claude Opus 5.5 (medium)24872433 to 254584.5$0.3101.9 minN/A#6
Claude Opus 5.5 (low)24602403 to 250984.3$0.2431.5 minN/A#7
MiMo-V2.6-Pro (thinking on)19201870 to 196580.2$0.0176.2 min10,176#45

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

Claude Opus 5.5 (max effort) scores higher on all nine metrics. Weighted by each metric's share of the Writing Score, Substance, Accuracy & Value does the most to put Claude Opus 5.5 (max effort) ahead, with a 7.1-point lead on that metric. The weight beside each metric is its share of the Writing Score.

Tone & Voice Match19% weight
82.8
78.0
Writing Craft & Clarity13% weight
82.9
77.8
Substance, Accuracy & Value15% weight
88.0
80.9
Continuity & Emotion14% weight
85.7
79.8
YouTube Best Practices12% weight
84.3
78.8
Hook Strength10% weight
83.7
79.7
Length Adherence8% weight
98.7
90.9
Slop Score5% weight
88.4
83.3
Visual Cue Quality4% weight
85.9
78.3

Consistency Across Drafts

Claude Opus 5.5 (max effort)MiMo-V2.6-Pro (thinking on)
Score spread across drafts (±)1.441.87
Lowest script score83.6 Script 1078.0 Script 4
Highest script score87.3 Script 782.5 Script 9
Range across scripts3.74.5

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 5.5 (max effort) minus MiMo-V2.6-Pro (thinking on), and the higher score is in bold.

ScriptClaude Opus 5.5 (max effort)MiMo-V2.6-Pro (thinking on)Gap
Script 1 opinion / warning explainer84.679.3+5.3
Script 2 news-analysis / skeptical explainer85.379.9+5.4
Script 3 personal roadmap / opinion86.579.7+6.8
Script 4 short explainer87.078.0+9.0
Script 5 news-analysis / opinion explainer85.981.5+4.4
Script 6 founder announcement / personal origin story86.381.1+5.2
Script 7 personal technical walkthrough / agentic workflow case study87.380.3+7.0
Script 8 product release announcement / personal observation85.580.7+4.8
Script 9 career guide / hiring analysis87.382.5+4.8
Script 10 engineering process walkthrough / presentation adaptation83.679.4+4.2

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