Model Comparison
DeepSeek V4.1 Flash 1017 vs Claude Haiku 5.5
DeepSeek V4.1 Flash 1017 (max) leads Claude Haiku 5.5 (max effort) by 99 Elo, and their 95% ranges sit 18 Elo apart. Among 81 models, each ranked by its best configuration, DeepSeek V4.1 Flash 1017 is #13 and Claude Haiku 5.5 is #17. Cost per script: $0.014 for DeepSeek V4.1 Flash 1017 (max) and $0.103 for Claude Haiku 5.5 (max effort).
Last updated , from board data generated at 16:59 UTC.
Picks and Verdict
PickDeepSeek V4.1 Flash 1017
DeepSeek V4.1 Flash 1017 (max)
Pick DeepSeek V4.1 Flash 1017 (max) as the everyday writer, since it leads Haiku's best by 99 Elo with separate ranges at $0.014 a script.
PickClaude Haiku 5.5
Claude Haiku 5.5 (xhigh)
Pick Claude Haiku 5.5 (xhigh) when substance and visual cues come first, since its 84.5 and 79.0 there beat both DeepSeek configurations for $0.029 a script.
The new DeepSeek snapshot pulls clear of Claude Haiku 5.5. DeepSeek V4.1 Flash 1017 (max) leads Claude Haiku 5.5 (max effort) by 99 Elo, and their 95% ranges sit 18 Elo apart, a small gap but enough to order them. Ranked by best configuration, DeepSeek is the #13 model of 81 and Haiku the #17 model. DeepSeek also costs $0.014 a script against $0.103 and takes 1.3 minutes against 12.6. One level down it gets lopsided. DeepSeek V4.1 Flash 1017 (high), under a cent, still overlaps Haiku's best, while Claude Haiku 5.5 (high) costs $0.012 and sits 347 Elo below it with separate ranges. Haiku needs max effort to compete at all, climbing 490 Elo from low at about 23 times the cost. Haiku's case is substance and polish: 3.3 points more on Substance, Accuracy & Value, 5.6 more on visual cues, plus a cleaner slop score and tighter length. DeepSeek takes tone, craft, continuity, YouTube Best Practices and hooks, by 1.7 to 3.3 points each. Claude Haiku 5.5 (xhigh) keeps most of Haiku's substance lead for $0.029. For everyday scripts I'd default to DeepSeek at max, and the open weights help.
Disclosure: Claude Haiku 5.5 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 DeepSeek V4.1 Flash 1017 and Claude Haiku 5.5, 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 |
|---|---|---|---|---|---|---|---|
| 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 |
| Claude Haiku 5.5 (max effort) | 1988 | 1950 to 2021 | 80.8 | $0.103 | 12.6 min | N/A | #33 |
| Claude Haiku 5.5 (xhigh) | 1916 | 1871 to 1953 | 80.0 | $0.029 | 3.6 min | N/A | #44 |
| Claude Haiku 5.5 (high) | 1703 | 1663 to 1741 | 78.0 | $0.012 | 1.5 min | N/A | #56 |
| Claude Haiku 5.5 (medium) | 1615 | 1555 to 1664 | 76.7 | $0.0067 | 0.8 min | N/A | #67 |
| Claude Haiku 5.5 (low) | 1498 | 1445 to 1537 | 75.2 | $0.0045 | 0.5 min | N/A | #79 |
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. DeepSeek V4.1 Flash 1017 (max) leads Claude Haiku 5.5 (max effort) by 99 Elo, and their 95% ranges do not overlap (2039 to 2140 against 1950 to 2021), which is strong evidence for that order.
- 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 Claude Haiku 5.5 (low) at Elo 1498 and $0.0045 per script by 552 Elo. Their 95% ranges do not overlap.
- Same Level Name, Side by Side. DeepSeek V4.1 Flash 1017 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 Configuration Overlapping the Top One. DeepSeek V4.1 Flash 1017 (high), at $0.0095 per script, is the cheapest configuration whose 95% range overlaps that of DeepSeek V4.1 Flash 1017 (max), at $0.014 per script: 2001 to 2092 against 2039 to 2140, so the board cannot separate them. DeepSeek V4.1 Flash 1017 (max) costs about 1.5 times as much per script.
- Cheapest DeepSeek V4.1 Flash 1017 Configuration Above Claude Haiku 5.5's Best. The cheapest DeepSeek V4.1 Flash 1017 configuration with a higher Writing Elo than Claude Haiku 5.5 (max effort), at Elo 1988 and $0.103 per script, is DeepSeek V4.1 Flash 1017 (high), at Elo 2050 and $0.0095 per script. Their 95% ranges overlap, so the board cannot separate them. It costs less than the configuration it beats.
- Lowest to Highest Thinking Level. DeepSeek V4.1 Flash 1017 moves +37 Elo from high to max, at about 1.5 times the cost per script. Claude Haiku 5.5 moves +490 Elo from low to max effort, at about 23 times the cost per script.
Writing Elo at Each Shared Level Name
| Level | DeepSeek V4.1 Flash 1017 | Claude Haiku 5.5 | Gap | 95% Ranges |
|---|---|---|---|---|
| high | 2050 | 1703 | +347 | Separate |
| max | 2087 | 1988 | +99 | Separate |
Gap is DeepSeek V4.1 Flash 1017 minus Claude Haiku 5.5 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
DeepSeek V4.1 Flash 1017 (max) scores higher on five of the nine metrics and Claude Haiku 5.5 (max effort) on four. Weighted by each metric's share of the Writing Score, Tone & Voice Match does the most to put DeepSeek V4.1 Flash 1017 (max) ahead, with a 2.8-point lead on that metric. The weight beside each metric is its share of the Writing Score.
Consistency Across Drafts
| DeepSeek V4.1 Flash 1017 (max) | Claude Haiku 5.5 (max effort) | |
|---|---|---|
| Score spread across drafts (±) | 1.58 | 1.55 |
| Lowest script score | 79.9 Script 4 | 78.2 Script 10 |
| Highest script score | 83.0 Script 7 | 82.6 Script 4 |
| Range across scripts | 3.1 | 4.4 |
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 DeepSeek V4.1 Flash 1017 (max) minus Claude Haiku 5.5 (max effort), and the higher score is in bold.
| Script | DeepSeek V4.1 Flash 1017 (max) | Claude Haiku 5.5 (max effort) | Gap |
|---|---|---|---|
| Script 1 opinion / warning explainer | 80.5 | 80.3 | +0.2 |
| Script 2 news-analysis / skeptical explainer | 80.8 | 80.7 | +0.1 |
| Script 3 personal roadmap / opinion | 80.5 | 80.8 | -0.3 |
| Script 4 short explainer | 79.9 | 82.6 | -2.7 |
| Script 5 news-analysis / opinion explainer | 82.9 | 81.7 | +1.2 |
| Script 6 founder announcement / personal origin story | 81.2 | 81.1 | +0.1 |
| Script 7 personal technical walkthrough / agentic workflow case study | 83.0 | 80.0 | +3.0 |
| Script 8 product release announcement / personal observation | 82.6 | 81.3 | +1.3 |
| Script 9 career guide / hiring analysis | 83.0 | 81.5 | +1.5 |
| Script 10 engineering process walkthrough / presentation adaptation | 81.6 | 78.2 | +3.4 |
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