Qwen3.8 Max vs DeepSeek V4 Flash 0731
Should you write with Qwen3.8 Max or DeepSeek V4 Flash 0731? On this page the only thing that changes is the model. Both wrote their own version of the scripts behind 10 of my What's AI videos, five drafts per script, and three judges from three model families scored every draft blind, with my finished script as the reference. DeepSeek V4 Flash 0731 leads overall, 84.4 to 83.2.
The verdict
Qwen3.8 Max costs roughly thirty times more per script here, and it's the weaker writer. DeepSeek V4 Flash 0731 ranks #59 against Qwen at #68, about 185 Elo ahead. The judges score it close to four points higher on both voice and continuity, the metrics for a script that sounds like a person talking, and it also has the stronger hook. And it costs about half a cent a script against $0.161, with open weights and less than half the wait. Qwen3.8 Max still has real strengths: it hits the target length far more reliably, scores better on slop and writes better visual cues. It's also the more consistent of the two from run to run, so if you need clean, on-length drafts every time and price doesn't matter, Qwen earns it.
Pick Qwen3.8 Max when you need clean, on-length drafts every time and the price doesn't matter.
Pick DeepSeek V4 Flash 0731 for better voice and flow at half a cent a script, with open weights.
How to read these numbers
Overall vs Elo. Overall is the weighted rubric score. Elo compares each model's five-draft average with every other model's, script by script, and counts gaps inside run-to-run noise as draws, so it rewards winning often and can disagree with overall. My own scripts score 90.9 on the same rubric, not a literal 100. Right now 2 of 167 ranked configurations reach it. A model above that line writes at that level on this rubric. That's not the same as beating my script line for line.
Reading the Elo intervals. Each 95% range comes from resampling every model's recorded run scores on each task and re-ranking. Comparing these separate ranges does not test the Elo gap between two models: far apart is strong evidence, and overlap isn't a tie. The tasks stay fixed, so the ranges say nothing about new tasks (the long version).
Metric by metric: where the gap comes from
Blue is Qwen3.8 Max, orange is DeepSeek V4 Flash 0731, and the solid bar takes the metric. The weight under each name is how much it counts toward the overall. Weighted by how much each metric counts, Tone & Voice Match does the most to put DeepSeek V4 Flash 0731 ahead: it leads there by 3.9 points. If you care about one thing, say voice or length, go straight to that row.
Cost, speed, and consistency
Quality is half the decision. The other half is what each script costs you, how long you wait, and how often you get a bad draft. DeepSeek V4 Flash 0731 is ahead and it's also the cheaper one: Qwen3.8 Max costs about 32x as much per script.
| Qwen3.8 Max | DeepSeek V4 Flash 0731 | |
|---|---|---|
| Overall / 100 | 83.2 | 84.4 |
| Writing Elo | 1682 | 1866 |
| Score spread across drafts (± overall std) | 1.980 | 3.360 |
| Cost per script (USD) | 0.161 | 0.005 |
| Avg time per script (min) | 7.2 | 3.1 |
| Open weights | No | Yes |
Green wins the row, and for spread, cost and time, lower wins. Cost is one script at recorded list prices, uncached and without judging, so your real bill will differ. Spread is how much the overall score moves across a model's drafts, and time is the average time to get one accepted script, retries included.
Want the script-by-script detail? Full scorecards: Qwen3.8 Max · DeepSeek V4 Flash 0731. The whole scoring pipeline is on the methodology page.
← All comparisons