Kimi K3 (thinking) vs DeepSeek V4 Pro (xhigh)
Should you write with Kimi K3 (thinking) or DeepSeek V4 Pro (xhigh)? 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. Kimi K3 (thinking) leads overall, 87.2 to 78.9.
The verdict
Kimi K3 (thinking), comfortably. It sits at #41, DeepSeek V4 Pro (xhigh) at #97, a gap of about 681 Elo, and Kimi takes all nine metrics. The judges mark DeepSeek down hardest on visual cues and YouTube structure, by eleven to thirteen points, the two things that turn a draft into a usable video script. Its overall score also swings more than twice as much between runs. Both are open weights, and DeepSeek is the cheaper one, about $0.013 a script against $0.095. If you want DeepSeek on this budget, though, other DeepSeek configurations on the board rank far higher than this xhigh setting.
Pick Kimi K3 (thinking) when you need usable cues and YouTube structure at a low price.
Pick DeepSeek V4 Pro (xhigh) only when you're committed to this exact route and cost is everything.
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 Kimi K3 (thinking), orange is DeepSeek V4 Pro (xhigh), 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 Kimi K3 (thinking) ahead: it leads there by 7.4 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. The lead has a price: Kimi K3 (thinking) costs about 7.3x as much per script as DeepSeek V4 Pro (xhigh).
| Kimi K3 (thinking) | DeepSeek V4 Pro (xhigh) | |
|---|---|---|
| Overall / 100 | 87.2 | 78.9 |
| Writing Elo | 2098 | 1417 |
| Score spread across drafts (± overall std) | 1.550 | 3.840 |
| Cost per script (USD) | 0.095 | 0.013 |
| Avg time per script (min) | 1.2 | 1.8 |
| Open weights | Yes | 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: Kimi K3 (thinking) · DeepSeek V4 Pro (xhigh). The whole scoring pipeline is on the methodology page.
← All comparisons