Towards AITowards AIToneBench

Lab Comparison

Alibaba (Qwen) Models Compared

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

All 8 ranked Alibaba (Qwen) 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.

Highest Writing Elo
Elo 1531Qwen3.8 Max (thinking on)#40 of 87 models · $0.207 per script
Lowest Cost Within 200 Elo of the Top
$0.017 per scriptQwen3.8 Flash (thinking on)Configuration #98 · Elo 1416
Lowest Cost
$0.0036 per scriptQwen3.5 9B (thinking on)Configuration #145 · Elo 147

A configuration 200 Elo below the top is expected to score about 24% in its script-by-script comparisons with it, a draw counting as half. The Lowest Cost Within 200 Elo of the Top card names the cheapest configuration that stays that close.

Recommended Configurations

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 pick Qwen3.8 Flash (thinking on). Qwen3.8 Max (thinking on) is 115 Elo ahead, and the ranges don't overlap, so the Max is the better writer. But it costs 12.2 times as much, $0.207 against $0.017, and for that gap I'd keep the money. One catch with both: they're slow. The Flash takes 6.8 minutes per script and the Max takes 11.2.

If you need speed, Qwen3.7 Max (high) comes back in 1.6 minutes, at #107. If you need open weights, the best open Qwen is down at #131, and Qwen3.5 9B (thinking on), the cheapest at $0.0036, is too far behind for me to use on a real script.

Qwen sits in the middle of the board at best. Qwen3.8 Max (thinking on) is the top Qwen configuration at #86, and its smaller sibling, Qwen3.8 Flash (thinking on), follows at #98. Go back one generation and Qwen3.7 Max (high) is already 298 Elo behind, and the open Qwen3 rows sit far down the board. So with Qwen, the generation you pick matters more than the size. Of the 8 configurations here, 5 are open weights, and I wouldn't write a script with any of those.

All 8 Alibaba (Qwen) Configurations, Ranked

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.

Score colors:82 and up65 to 81.9under 65

With the metric scores shown, the table scrolls sideways.

Configuration RankConfigurationWriting EloWriting ScoreCost / ScriptTime / ScriptTone & Voice MatchWriting Craft & ClaritySubstance, Accuracy & ValueHook StrengthYouTube Best PracticesContinuity & EmotionSlop ScoreLength AdherenceVisual Cue Quality
#86Qwen3.8 Max (thinking on)1531 ±3776.8$0.20711.2 min70.370.982.079.274.074.080.297.574.8
#98Qwen3.8 Flash (thinking on)1416 ±4375.3$0.0176.8 min69.868.679.777.272.972.778.497.170.4
#107Qwen3.7 Max (high)1233 ±4872.6$0.1021.6 min64.169.574.775.173.473.278.787.067.0
#131Qwen3 235B Thinking (thinking always on) · open weights602 ±4362.6$0.0121.4 min54.462.559.072.263.864.357.179.856.0
#132Qwen3 235B A22B (no thinking) · open weights599 ±3462.4$0.00571.0 min58.557.966.071.258.363.955.476.248.3
#144Qwen3 Next 80B (no thinking) · open weights176 ±4452.3$0.00480.2 min49.743.855.068.645.256.053.762.328.6
#145Qwen3.5 9B (thinking on) · open weights147 ±7550.2$0.00360.9 min46.542.551.162.948.047.469.056.836.7
#147Qwen3 8B (thinking on) · open weights6 ±3947.1$0.00670.9 min42.149.751.362.840.244.862.028.156.9

Qwen3.8 Max Against Other Alibaba (Qwen) Models

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.

← All comparisons