DeepSeek V4.1 Flash (max) vs Qwen3.8 Flash

Head-to-head on the Towards AI writing benchmark: same 10 real YouTube scripts, five runs each, scored blind by a three-family judge panel. DeepSeek V4.1 Flash (max) leads overall, 88.1 to 83.0.

DeepSeek V4.1 Flash (max)

#28 Elo 2183 · 88.1/100

Qwen3.8 Flash

#65 Elo 1699 · 83.0/100

Cost / script

$0.013 vs $0.013

Human baseline

90.9 DeepSeek V4.1 Flash (max) falls below it · Qwen3.8 Flash falls below it

The verdict

DeepSeek V4.1 Flash (max) wins this one, and the board is not close: #28 at 2183.1 Elo and 88.11 overall, against #65, 1698.8 Elo and 82.97 overall for Qwen3.8 Flash. The 95% Elo intervals do not overlap (2147.5 to 2214.6 against 1650.3 to 1754.4). Where DeepSeek V4.1 Flash (max) pulls ahead: Continuity & Emotion (86.86 vs 78.74) and Writing Craft & Clarity (88.82 vs 82.19). Qwen3.8 Flash still wins on Length Adherence (92.46 vs 91.98), so it is not a clean sweep. The trade is price: DeepSeek V4.1 Flash (max) costs about $0.013 per article against $0.013 for Qwen3.8 Flash, so you pay less for the higher rank. DeepSeek V4.1 Flash (max) publishes open weights; Qwen3.8 Flash does not. Default to DeepSeek V4.1 Flash (max). Reach for Qwen3.8 Flash when length adherence matters more to you than the headline rank.

Pick DeepSeek V4.1 Flash (max) for the stronger board result, continuity, and writing craft.
Pick Qwen3.8 Flash for length adherence.

Reading the Elo intervals. Each range estimates uncertainty around one model's Elo. We repeatedly sample from every model's recorded run scores on each task and recalculate the full ranking. Comparing these separate ranges does not test the Elo gap between two models. The tasks stay fixed, so the ranges do not measure how the ranking would change on new tasks.

Metric by metric

Blue bars: DeepSeek V4.1 Flash (max). Orange bars: Qwen3.8 Flash. Same 0–100 scale; the bold bar wins that metric.

Tone & Voice Match19% weight

88.0

81.9

Writing Craft & Clarity13% weight

88.8

82.2

Substance, Accuracy & Value15% weight

86.8

83.2

Continuity & Emotion14% weight

86.9

78.7

YouTube Best Practices12% weight

87.8

82.2

Hook Strength10% weight

88.3

83.5

Length Adherence8% weight

92.0

92.5

Slop Score (EQ-Bench + ours)5% weight

91.3

88.6

Visual Cue Quality4% weight

84.5

79.5

Everything else that differs

	DeepSeek V4.1 Flash (max)	Qwen3.8 Flash
Overall / 100	88.1	83.0
Writing Elo	2183	1699
Run-to-run spread (± overall std)	1.120	2.420
Cost per script (USD)	0.013	0.013
Avg latency (s)	76.2	341.7
Open weights	Yes	No

Full scorecards: DeepSeek V4.1 Flash (max) · Qwen3.8 Flash. How scoring works: methodology.

← All comparisons