Model Comparison
Claude Sonnet 5.5 vs Kimi K3
Claude Sonnet 5.5 (max effort) leads Kimi K3 (max) by 358 Elo, and their 95% ranges sit 261 Elo apart. Among 78 models, each ranked by its best configuration, Claude Sonnet 5.5 is #2 and Kimi K3 is #10. Cost per script: $1.38 for Claude Sonnet 5.5 (max effort) and $0.362 for Kimi K3 (max).
Last updated , from board data generated at 01:13 UTC.
Picks and Verdict
PickClaude Sonnet 5.5
Claude Sonnet 5.5 (xhigh)
Pick Claude Sonnet 5.5 (xhigh) over Kimi K3 (max), since it ranks higher with separate ranges and costs less, at $0.302 a script.
PickKimi K3
Kimi K3 (high)
Pick Kimi K3 (high) for open weights and a quick $0.128 draft that clearly beats Sonnet at its lowest setting.
Against Kimi K3, the Claude Sonnet 5.5 setting I'd look at first is xhigh. At $0.302 a script it costs less than Kimi K3 (max) at $0.362, finishes in about a third of the time, and still sits 145 Elo higher with ranges that don't overlap. Sonnet at max effort pulls further ahead, 358 Elo up, at close to four times Kimi's price, and the board separates it from xhigh too, so that extra money buys something. Ranked by best setting, Sonnet is the #2 model of 78 and Kimi K3 the #10 model. Lower down, the two get close. At the level both call high, the board can't separate them, and Sonnet's high is the cheaper and faster one at $0.092 and about a minute. Kimi at high does beat Sonnet at low with separate ranges, though, so if you were going to run Sonnet at its cheapest, Kimi is the better draft for about twice the price. Kimi also writes stronger hooks than Sonnet's top two settings, and its weights are open if you want to host it yourself. Sonnet at xhigh is my pick here.
Disclosure: Claude Sonnet 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 Claude Sonnet 5.5 and Kimi K3, 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 |
|---|---|---|---|---|---|---|---|
| Claude Sonnet 5.5 (max effort) | 2561 | 2504 to 2614 | 84.6 | $1.38 | 14.3 min | N/A | #3 |
| Claude Sonnet 5.5 (xhigh) | 2348 | 2304 to 2383 | 83.2 | $0.302 | 3.1 min | N/A | #10 |
| Kimi K3 (max) | 2203 | 2158 to 2243 | 82.2 | $0.362 | 9.9 min | 18,568 | #20 |
| Claude Sonnet 5.5 (high) | 2040 | 1984 to 2087 | 81.0 | $0.092 | 1.0 min | N/A | #27 |
| Kimi K3 (high) | 1978 | 1936 to 2029 | 80.3 | $0.128 | 2.8 min | 3,081 | #37 |
| Claude Sonnet 5.5 (medium) | 1856 | 1812 to 1900 | 79.4 | $0.065 | 0.6 min | N/A | #47 |
| Claude Sonnet 5.5 (low) | 1808 | 1771 to 1843 | 78.9 | $0.061 | 0.7 min | N/A | #50 |
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. Claude Sonnet 5.5 (max effort) leads Kimi K3 (max) by 358 Elo, and their 95% ranges do not overlap (2504 to 2614 against 2158 to 2243), which is strong evidence for that order.
- Lowest Thinking Levels Head to Head. At their lowest thinking levels, Kimi K3 (high) at Elo 1978 and $0.128 per script leads Claude Sonnet 5.5 (low) at Elo 1808 and $0.061 per script by 170 Elo. Their 95% ranges do not overlap.
- Same Level Name, Side by Side. Claude Sonnet 5.5 has the higher Writing Elo at every level name both models use: high and max. The 95% ranges do not overlap at max. Level names come from each provider, so the same name does not promise the same reasoning budget.
- Cheapest Claude Sonnet 5.5 Configuration Above Kimi K3's Best. The cheapest Claude Sonnet 5.5 configuration with a higher Writing Elo than Kimi K3 (max), at Elo 2203 and $0.362 per script, is Claude Sonnet 5.5 (xhigh), at Elo 2348 and $0.302 per script. It costs less than the configuration it beats.
- Lowest to Highest Thinking Level. Claude Sonnet 5.5 moves +753 Elo from low to max effort, at about 23 times the cost per script. Kimi K3 moves +225 Elo from high to max, at about 2.8 times the cost per script.
Writing Elo at Each Shared Level Name
| Level | Claude Sonnet 5.5 | Kimi K3 | Gap | 95% Ranges |
|---|---|---|---|---|
| high | 2040 | 1978 | +62 | Overlap |
| max | 2561 | 2203 | +358 | Separate |
Gap is Claude Sonnet 5.5 minus Kimi K3 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 Sonnet 5.5 (max effort) scores higher on eight of the nine metrics and Kimi K3 (max) on one. Weighted by each metric's share of the Writing Score, Substance, Accuracy & Value does the most to put Claude Sonnet 5.5 (max effort) ahead, with a 4.3-point lead on that metric. The weight beside each metric is its share of the Writing Score.
Consistency Across Drafts
| Claude Sonnet 5.5 (max effort) | Kimi K3 (max) | |
|---|---|---|
| Score spread across drafts (±) | 1.22 | 1.68 |
| Lowest script score | 83.0 Script 10 | 80.0 Script 10 |
| Highest script score | 86.7 Script 9 | 83.8 Script 9 |
| Range across scripts | 3.7 | 3.8 |
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 Sonnet 5.5 (max effort) minus Kimi K3 (max), and the higher score is in bold.
| Script | Claude Sonnet 5.5 (max effort) | Kimi K3 (max) | Gap |
|---|---|---|---|
| Script 1 opinion / warning explainer | 84.2 | 81.4 | +2.8 |
| Script 2 news-analysis / skeptical explainer | 84.8 | 82.6 | +2.2 |
| Script 3 personal roadmap / opinion | 84.0 | 81.8 | +2.2 |
| Script 4 short explainer | 85.0 | 80.1 | +4.9 |
| Script 5 news-analysis / opinion explainer | 84.8 | 83.4 | +1.4 |
| Script 6 founder announcement / personal origin story | 84.6 | 83.2 | +1.4 |
| Script 7 personal technical walkthrough / agentic workflow case study | 85.0 | 83.6 | +1.4 |
| Script 8 product release announcement / personal observation | 83.9 | 82.3 | +1.6 |
| Script 9 career guide / hiring analysis | 86.7 | 83.8 | +2.9 |
| Script 10 engineering process walkthrough / presentation adaptation | 83.0 | 80.0 | +3.0 |
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