Model Comparison
GPT-6 Astra vs DeepSeek V4.1 Flash 1017
Too close to call: the best configurations, GPT-6 Astra (max) and DeepSeek V4.1 Flash 1017 (max), are 72 Elo apart, and their 95% ranges overlap. Among 81 models, each ranked by its best configuration, GPT-6 Astra is #11 and DeepSeek V4.1 Flash 1017 is #13. Cost per script: $1.15 for GPT-6 Astra (max) and $0.014 for DeepSeek V4.1 Flash 1017 (max).
Last updated , from board data generated at 16:59 UTC.
Picks and Verdict
PickGPT-6 Astra
GPT-6 Astra (max)
Pick GPT-6 Astra (max) for a fact-heavy technical script, since it leads DeepSeek's best by 5.2 points on substance and 8.1 on visual cues, at $1.15 a script.
PickDeepSeek V4.1 Flash 1017
DeepSeek V4.1 Flash 1017 (max)
Pick DeepSeek V4.1 Flash 1017 (max) for nearly everything else, since the board can't separate it from Astra's best and it costs $0.014 a script.
Here the price gap is hard to justify: the board can't separate GPT-6 Astra's best from DeepSeek V4.1 Flash 1017 at max, and Astra costs about 82 times as much. GPT-6 Astra (max) sits 72 Elo above DeepSeek V4.1 Flash 1017 (max), and their 95% ranges overlap. Ranked by best configuration, Astra is the #11 model of 81 and DeepSeek the #13 model. Astra at max costs $1.15 and 9.2 minutes a script, DeepSeek $0.014 and 1.3 minutes. Astra's case is facts and screen work, 8.1 points ahead on visual cues and 5.2 on substance, with a cleaner slop score. DeepSeek leads YouTube Best Practices by 3.5 and is ahead on tone, craft and continuity, with the steadiest drafts on this page. The lower Astra level is the one I'd skip. GPT-6 Astra (medium) costs $0.262 and lands 105 Elo below DeepSeek at max with separate ranges, so on Astra the 177 Elo jump to max is the only step worth paying for. I'd keep Astra at max for a technical script that lives on its facts and send everything else to DeepSeek at max.
Disclosure: GPT-6 Astra comes from the same lab as one of the four judges, GPT-6.1 Sol. 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 GPT-6 Astra and DeepSeek V4.1 Flash 1017, scored on the same 10 scripts by the same four judges.
- 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.
- When two bars overlap, the board cannot order those two configurations. Bars far apart are strong evidence for the order.
- 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).
Shaded rows are the configurations chosen for the metric comparison below.
| Configuration | Writing Elo | 95% Range | Writing Score | Cost / Script | Time / Script | Reasoning Tokens | Configuration Rank |
|---|---|---|---|---|---|---|---|
| GPT-6 Astra (max) | 2159 | 2118 to 2199 | 82.3 | $1.15 | 9.2 min | N/A | #21 |
| DeepSeek V4.1 Flash 1017 (max) | 2087 | 2039 to 2140 | 81.6 | $0.014 | 1.3 min | 16,960 | #24 |
| DeepSeek V4.1 Flash 1017 (high) | 2050 | 2001 to 2092 | 81.1 | $0.0095 | 0.9 min | 9,728 | #27 |
| GPT-6 Astra (medium) | 1982 | 1952 to 2014 | 80.7 | $0.262 | 1.5 min | N/A | #36 |
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. Where two 95% ranges overlap, the board cannot order those two configurations.
- Best Configurations Head to Head. GPT-6 Astra (max) at Elo 2159 (range 2118 to 2199) and DeepSeek V4.1 Flash 1017 (max) at Elo 2087 (range 2039 to 2140) have overlapping 95% ranges, so the board cannot separate the two best configurations.
- Lowest Thinking Levels Head to Head. At their lowest thinking levels, DeepSeek V4.1 Flash 1017 (high) at Elo 2050 and $0.0095 per script leads GPT-6 Astra (medium) at Elo 1982 and $0.262 per script by 68 Elo. Their 95% ranges overlap, so the board cannot separate them.
- Same Level Name, Side by Side. At max, the one level name both models use, GPT-6 Astra has the higher Writing Elo: 2159 against 2087. Their 95% ranges overlap there. Level names come from each provider, so the same name does not promise the same reasoning budget.
- Cheapest Configuration Overlapping the Top One. DeepSeek V4.1 Flash 1017 (max), at $0.014 per script, is the cheapest configuration whose 95% range overlaps that of GPT-6 Astra (max), at $1.15 per script: 2039 to 2140 against 2118 to 2199, so the board cannot separate them. GPT-6 Astra (max) costs about 82 times as much per script.
- Cheapest GPT-6 Astra Configuration Above DeepSeek V4.1 Flash 1017's Best. The cheapest GPT-6 Astra configuration with a higher Writing Elo than DeepSeek V4.1 Flash 1017 (max), at Elo 2087 and $0.014 per script, is GPT-6 Astra (max), at Elo 2159 and $1.15 per script. Their 95% ranges overlap, so the board cannot separate them. It costs about 82 times as much per script.
- Lowest to Highest Thinking Level. GPT-6 Astra moves +177 Elo from medium to max, at about 4.4 times the cost per script. DeepSeek V4.1 Flash 1017 moves +37 Elo from high to max, at about 1.5 times the cost per script.
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
GPT-6 Astra (max) scores higher on five of the nine metrics and DeepSeek V4.1 Flash 1017 (max) on four. Weighted by each metric's share of the Writing Score, Substance, Accuracy & Value does the most to put GPT-6 Astra (max) ahead, with a 5.2-point lead on that metric. The weight beside each metric is its share of the Writing Score.
Consistency Across Drafts
| GPT-6 Astra (max) | DeepSeek V4.1 Flash 1017 (max) | |
|---|---|---|
| Score spread across drafts (±) | 1.87 | 1.58 |
| Lowest script score | 80.6 Script 1 | 79.9 Script 4 |
| Highest script score | 85.8 Script 9 | 83.0 Script 7 |
| Range across scripts | 5.2 | 3.1 |
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 GPT-6 Astra (max) minus DeepSeek V4.1 Flash 1017 (max), and the higher score is in bold.
| Script | GPT-6 Astra (max) | DeepSeek V4.1 Flash 1017 (max) | Gap |
|---|---|---|---|
| Script 1 opinion / warning explainer | 80.6 | 80.5 | +0.1 |
| Script 2 news-analysis / skeptical explainer | 82.0 | 80.8 | +1.2 |
| Script 3 personal roadmap / opinion | 84.3 | 80.5 | +3.8 |
| Script 4 short explainer | 82.7 | 79.9 | +2.8 |
| Script 5 news-analysis / opinion explainer | 81.3 | 82.9 | -1.6 |
| Script 6 founder announcement / personal origin story | 83.6 | 81.2 | +2.4 |
| Script 7 personal technical walkthrough / agentic workflow case study | 81.6 | 83.0 | -1.4 |
| Script 8 product release announcement / personal observation | 80.6 | 82.6 | -2.0 |
| Script 9 career guide / hiring analysis | 85.8 | 83.0 | +2.8 |
| Script 10 engineering process walkthrough / presentation adaptation | 80.7 | 81.6 | -0.9 |
Scorecards and Thinking Pages
Each configuration name in the table above opens its own scorecard.
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