Model Comparison
Claude Fable 5.2 vs DeepSeek V4.1 Flash 1017
Claude Fable 5.2 (xhigh) leads DeepSeek V4.1 Flash 1017 (max) by 649 Elo, and their 95% ranges sit 564 Elo apart. Even Claude Fable 5.2's lowest-rated configuration, Claude Fable 5.2 (low), ranks above DeepSeek V4.1 Flash 1017 (max). Among 87 models, each ranked by its best configuration, Claude Fable 5.2 is #2 and DeepSeek V4.1 Flash 1017 is #15. Cost per script: $2.22 for Claude Fable 5.2 (xhigh) and $0.014 for DeepSeek V4.1 Flash 1017 (max).
Last updated , from board data generated at 17:52 UTC.
Picks and Verdict
PickClaude Fable 5.2
Claude Fable 5.2 (xhigh)
Pick Claude Fable 5.2 (xhigh) for the script you'll record, since it leads DeepSeek's best by 649 Elo and comes out ahead on all nine metrics.
PickDeepSeek V4.1 Flash 1017
DeepSeek V4.1 Flash 1017 (max)
Pick DeepSeek V4.1 Flash 1017 (max) for open-weights drafting in bulk, at $0.014 a script and 1.3 minutes.
The newer DeepSeek snapshot doesn't get anywhere near Claude Fable 5.2. Fable 5.2 is the #2 model of 87 and DeepSeek V4.1 Flash 1017 the #15 model, each ranked by its best configuration. Claude Fable 5.2 (xhigh) leads DeepSeek V4.1 Flash 1017 (max) by 649 Elo, with 95% ranges 564 Elo apart, and even Claude Fable 5.2 (low) sits 391 Elo above DeepSeek's best, at about the same speed, 1.4 minutes against 1.3. At the top, Fable leads all nine metrics, by 10.0 points on Visual Cue Quality and 5.9 on substance, and its drafts are steadier than either DeepSeek setting's. Then you see the bill: $2.22 and 7.4 minutes a script at xhigh against $0.014 and 1.3 minutes, and even Fable at low costs about 33 times DeepSeek at max. The thinking ladders read differently too. Fable gains 162 Elo from high to xhigh with separate ranges, so on Fable that step is worth paying for, while max effort costs about 2.6 times xhigh's price for 27 Elo less. DeepSeek gains 50 from high to max inside overlapping ranges, for under half a cent more, and max lands closer to the target length, so I'd leave it on max. Fable 5.2 for the script I record, DeepSeek for drafting in bulk.
Disclosure: Claude Fable 5.2 comes from the same lab as one of the four judges, Claude Opus 5.5. 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 Claude Fable 5.2 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.
- Each bar describes one configuration. Overlap alone does not establish a tie or decide their order. Comparing these ranges does not directly test the difference between a pair.
- 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 |
|---|---|---|---|---|---|---|---|
| Claude Fable 5.2 (xhigh) | 2673 | 2630 to 2717 | 85.6 | $2.22 | 7.4 min | N/A | #3 |
| Claude Fable 5.2 (max effort) | 2646 | 2599 to 2689 | 85.5 | $5.75 | 19.9 min | N/A | #4 |
| Claude Fable 5.2 (high) | 2511 | 2455 to 2555 | 84.8 | $1.22 | 4.0 min | N/A | #5 |
| Claude Fable 5.2 (medium) | 2422 | 2370 to 2469 | 84.3 | $0.561 | 1.8 min | N/A | #11 |
| Claude Fable 5.2 (low) | 2415 | 2370 to 2454 | 84.1 | $0.460 | 1.4 min | N/A | #12 |
| DeepSeek V4.1 Flash 1017 (max) | 2024 | 1979 to 2066 | 81.6 | $0.014 | 1.3 min | 16,960 | #30 |
| DeepSeek V4.1 Flash 1017 (high) | 1974 | 1929 to 2021 | 81.1 | $0.0095 | 0.9 min | 9,728 | #34 |
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. The 95% ranges describe individual configurations; overlap alone establishes neither a tie nor an order for a pair.
- Best Configurations Head to Head. Claude Fable 5.2 (xhigh) leads DeepSeek V4.1 Flash 1017 (max) by 649 Elo, and their 95% ranges do not overlap (2630 to 2717 against 1979 to 2066), which is strong evidence for that order.
- Every Claude Fable 5.2 Configuration Ranks Higher. All five Claude Fable 5.2 configurations rank above both DeepSeek V4.1 Flash 1017 configurations: the lowest-rated, Claude Fable 5.2 (low) at Elo 2415, still beats DeepSeek V4.1 Flash 1017 (max) at Elo 2024. Their 95% ranges do not overlap.
- Lowest Thinking Levels Head to Head. At their lowest thinking levels, Claude Fable 5.2 (low) at Elo 2415 and $0.460 per script leads DeepSeek V4.1 Flash 1017 (high) at Elo 1974 and $0.0095 per script by 441 Elo. Their 95% ranges do not overlap.
- Same Level Name, Side by Side. Claude Fable 5.2 has the higher Writing Elo at every level name both models use: high and max. The 95% ranges do not overlap at any of them. Level names come from each provider, so the same name does not promise the same reasoning budget.
- Cheapest Claude Fable 5.2 Configuration Above DeepSeek V4.1 Flash 1017's Best. The cheapest Claude Fable 5.2 configuration with a higher Writing Elo than DeepSeek V4.1 Flash 1017 (max), at Elo 2024 and $0.014 per script, is Claude Fable 5.2 (low), at Elo 2415 and $0.460 per script. It costs about 33 times as much per script.
- Lowest to Highest Thinking Level. Claude Fable 5.2 moves +231 Elo from low to max effort, at about 13 times the cost per script. DeepSeek V4.1 Flash 1017 moves +50 Elo from high to max, at about 1.5 times the cost per script.
Writing Elo at Each Shared Level Name
| Level | Claude Fable 5.2 | DeepSeek V4.1 Flash 1017 | Gap | 95% Ranges |
|---|---|---|---|---|
| high | 2511 | 1974 | +537 | Separate |
| max | 2646 | 2024 | +622 | Separate |
Gap is Claude Fable 5.2 minus DeepSeek V4.1 Flash 1017 at the same level name, using each model's highest-rated configuration at that level.
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
Claude Fable 5.2 (xhigh) scores higher on all nine metrics. Weighted by each metric's share of the Writing Score, Substance, Accuracy & Value does the most to put Claude Fable 5.2 (xhigh) ahead, with a 5.9-point lead on that metric. The weight beside each metric is its share of the Writing Score.
Consistency Across Drafts
| Claude Fable 5.2 (xhigh) | DeepSeek V4.1 Flash 1017 (max) | |
|---|---|---|
| Score spread across drafts (±) | 1.44 | 1.58 |
| Lowest script score | 83.6 Script 10 | 79.9 Script 4 |
| Highest script score | 88.0 Script 9 | 83.0 Script 7 |
| Range across scripts | 4.4 | 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 Claude Fable 5.2 (xhigh) minus DeepSeek V4.1 Flash 1017 (max), and the higher score is in bold.
| Script | Claude Fable 5.2 (xhigh) | DeepSeek V4.1 Flash 1017 (max) | Gap |
|---|---|---|---|
| Script 1 opinion / warning explainer | 83.7 | 80.5 | +3.2 |
| Script 2 news-analysis / skeptical explainer | 85.4 | 80.8 | +4.6 |
| Script 3 personal roadmap / opinion | 85.5 | 80.5 | +5.0 |
| Script 4 short explainer | 85.4 | 79.9 | +5.5 |
| Script 5 news-analysis / opinion explainer | 86.1 | 82.9 | +3.2 |
| Script 6 founder announcement / personal origin story | 85.9 | 81.2 | +4.7 |
| Script 7 personal technical walkthrough / agentic workflow case study | 87.3 | 83.0 | +4.3 |
| Script 8 product release announcement / personal observation | 85.1 | 82.6 | +2.5 |
| Script 9 career guide / hiring analysis | 88.0 | 83.0 | +5.0 |
| Script 10 engineering process walkthrough / presentation adaptation | 83.6 | 81.6 | +2.0 |
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