Skip to the comparison

Compare 14 AI coding setups, two at a time

Each page puts two leaderboard setups side by side on the same 24 tasks: Stet score, cost, agent time and which tasks each solved.

Data from See all 14 setups on the leaderboard

Stet score, each row against each column
1SWE-2 Max2GPT 6.1 Sol medium3Opus 54Grok 4.7 High5Pareto 26.96Haiku 5.5 max7GPT 6 Sol medium8GPT 6 Luna max9DeepSeek V4.1 Flash10Haiku 5.5 low11Opus 4.812Muse control13GLM 5.3 Flash14GPT 6 Luna low
1SWE-2 MaxSWE-2 Max vs GPT 6.1 Sol medium, Leads, ranges overlapSWE-2 Max vs Opus 5, Leads, ranges overlapSWE-2 Max vs Grok 4.7 High, Leads, ranges overlapSWE-2 Max vs Pareto 26.9, Leads, ranges overlapSWE-2 Max vs Haiku 5.5 max, Leads, ranges overlapSWE-2 Max vs GPT 6 Sol medium, Leads, ranges overlapSWE-2 Max vs GPT 6 Luna max, Leads, ranges overlapSWE-2 Max vs DeepSeek V4.1 Flash, Leads, ranges overlapSWE-2 Max vs Haiku 5.5 low, Leads, 95% ranges apartSWE-2 Max vs Opus 4.8, Leads, 95% ranges apartSWE-2 Max vs Muse control, Leads, 95% ranges apartSWE-2 Max vs GLM 5.3 Flash, Leads, 95% ranges apartSWE-2 Max vs GPT 6 Luna low, Leads, 95% ranges apart
2GPT 6.1 Sol mediumGPT 6.1 Sol medium vs SWE-2 Max, Trails, ranges overlapGPT 6.1 Sol medium vs Opus 5, Leads, ranges overlapGPT 6.1 Sol medium vs Grok 4.7 High, Leads, ranges overlapGPT 6.1 Sol medium vs Pareto 26.9, Leads, ranges overlapGPT 6.1 Sol medium vs Haiku 5.5 max, Leads, ranges overlapGPT 6.1 Sol medium vs GPT 6 Sol medium, Leads, ranges overlapGPT 6.1 Sol medium vs GPT 6 Luna max, Leads, ranges overlapGPT 6.1 Sol medium vs DeepSeek V4.1 Flash, Leads, ranges overlapGPT 6.1 Sol medium vs Haiku 5.5 low, Leads, 95% ranges apartGPT 6.1 Sol medium vs Opus 4.8, Leads, 95% ranges apartGPT 6.1 Sol medium vs Muse control, Leads, 95% ranges apartGPT 6.1 Sol medium vs GLM 5.3 Flash, Leads, 95% ranges apartGPT 6.1 Sol medium vs GPT 6 Luna low, Leads, 95% ranges apart
3Opus 5Opus 5 vs SWE-2 Max, Trails, ranges overlapOpus 5 vs GPT 6.1 Sol medium, Trails, ranges overlapOpus 5 vs Grok 4.7 High, Leads, ranges overlapOpus 5 vs Pareto 26.9, Leads, ranges overlapOpus 5 vs Haiku 5.5 max, Leads, ranges overlapOpus 5 vs GPT 6 Sol medium, Leads, ranges overlapOpus 5 vs GPT 6 Luna max, Leads, ranges overlapOpus 5 vs DeepSeek V4.1 Flash, Leads, ranges overlapOpus 5 vs Haiku 5.5 low, Leads, 95% ranges apartOpus 5 vs Opus 4.8, Leads, 95% ranges apartOpus 5 vs Muse control, Leads, 95% ranges apartOpus 5 vs GLM 5.3 Flash, Leads, 95% ranges apartOpus 5 vs GPT 6 Luna low, Leads, 95% ranges apart
4Grok 4.7 HighGrok 4.7 High vs SWE-2 Max, Trails, ranges overlapGrok 4.7 High vs GPT 6.1 Sol medium, Trails, ranges overlapGrok 4.7 High vs Opus 5, Trails, ranges overlapGrok 4.7 High vs Pareto 26.9, Leads, ranges overlapGrok 4.7 High vs Haiku 5.5 max, Leads, ranges overlapGrok 4.7 High vs GPT 6 Sol medium, Leads, ranges overlapGrok 4.7 High vs GPT 6 Luna max, Leads, ranges overlapGrok 4.7 High vs DeepSeek V4.1 Flash, Leads, ranges overlapGrok 4.7 High vs Haiku 5.5 low, Leads, 95% ranges apartGrok 4.7 High vs Opus 4.8, Leads, 95% ranges apartGrok 4.7 High vs Muse control, Leads, 95% ranges apartGrok 4.7 High vs GLM 5.3 Flash, Leads, 95% ranges apartGrok 4.7 High vs GPT 6 Luna low, Leads, 95% ranges apart
5Pareto 26.9Pareto 26.9 vs SWE-2 Max, Trails, ranges overlapPareto 26.9 vs GPT 6.1 Sol medium, Trails, ranges overlapPareto 26.9 vs Opus 5, Trails, ranges overlapPareto 26.9 vs Grok 4.7 High, Trails, ranges overlapPareto 26.9 vs Haiku 5.5 max, Leads, ranges overlapPareto 26.9 vs GPT 6 Sol medium, Leads, ranges overlapPareto 26.9 vs GPT 6 Luna max, Leads, ranges overlapPareto 26.9 vs DeepSeek V4.1 Flash, Leads, ranges overlapPareto 26.9 vs Haiku 5.5 low, Leads, 95% ranges apartPareto 26.9 vs Opus 4.8, Leads, ranges overlapPareto 26.9 vs Muse control, Leads, 95% ranges apartPareto 26.9 vs GLM 5.3 Flash, Leads, 95% ranges apartPareto 26.9 vs GPT 6 Luna low, Leads, 95% ranges apart
6Haiku 5.5 maxHaiku 5.5 max vs SWE-2 Max, Trails, ranges overlapHaiku 5.5 max vs GPT 6.1 Sol medium, Trails, ranges overlapHaiku 5.5 max vs Opus 5, Trails, ranges overlapHaiku 5.5 max vs Grok 4.7 High, Trails, ranges overlapHaiku 5.5 max vs Pareto 26.9, Trails, ranges overlapHaiku 5.5 max vs GPT 6 Sol medium, Leads, ranges overlapHaiku 5.5 max vs GPT 6 Luna max, Leads, ranges overlapHaiku 5.5 max vs DeepSeek V4.1 Flash, Leads, ranges overlapHaiku 5.5 max vs Haiku 5.5 low, Leads, ranges overlapHaiku 5.5 max vs Opus 4.8, Leads, ranges overlapHaiku 5.5 max vs Muse control, Leads, 95% ranges apartHaiku 5.5 max vs GLM 5.3 Flash, Leads, 95% ranges apartHaiku 5.5 max vs GPT 6 Luna low, Leads, 95% ranges apart
7GPT 6 Sol mediumGPT 6 Sol medium vs SWE-2 Max, Trails, ranges overlapGPT 6 Sol medium vs GPT 6.1 Sol medium, Trails, ranges overlapGPT 6 Sol medium vs Opus 5, Trails, ranges overlapGPT 6 Sol medium vs Grok 4.7 High, Trails, ranges overlapGPT 6 Sol medium vs Pareto 26.9, Trails, ranges overlapGPT 6 Sol medium vs Haiku 5.5 max, Trails, ranges overlapGPT 6 Sol medium vs GPT 6 Luna max, Leads, ranges overlapGPT 6 Sol medium vs DeepSeek V4.1 Flash, Leads, ranges overlapGPT 6 Sol medium vs Haiku 5.5 low, Leads, 95% ranges apartGPT 6 Sol medium vs Opus 4.8, Leads, ranges overlapGPT 6 Sol medium vs Muse control, Leads, 95% ranges apartGPT 6 Sol medium vs GLM 5.3 Flash, Leads, 95% ranges apartGPT 6 Sol medium vs GPT 6 Luna low, Leads, 95% ranges apart
8GPT 6 Luna maxGPT 6 Luna max vs SWE-2 Max, Trails, ranges overlapGPT 6 Luna max vs GPT 6.1 Sol medium, Trails, ranges overlapGPT 6 Luna max vs Opus 5, Trails, ranges overlapGPT 6 Luna max vs Grok 4.7 High, Trails, ranges overlapGPT 6 Luna max vs Pareto 26.9, Trails, ranges overlapGPT 6 Luna max vs Haiku 5.5 max, Trails, ranges overlapGPT 6 Luna max vs GPT 6 Sol medium, Trails, ranges overlapGPT 6 Luna max vs DeepSeek V4.1 Flash, Leads, ranges overlapGPT 6 Luna max vs Haiku 5.5 low, Leads, ranges overlapGPT 6 Luna max vs Opus 4.8, Leads, ranges overlapGPT 6 Luna max vs Muse control, Leads, ranges overlapGPT 6 Luna max vs GLM 5.3 Flash, Leads, ranges overlapGPT 6 Luna max vs GPT 6 Luna low, Leads, 95% ranges apart
9DeepSeek V4.1 FlashDeepSeek V4.1 Flash vs SWE-2 Max, Trails, ranges overlapDeepSeek V4.1 Flash vs GPT 6.1 Sol medium, Trails, ranges overlapDeepSeek V4.1 Flash vs Opus 5, Trails, ranges overlapDeepSeek V4.1 Flash vs Grok 4.7 High, Trails, ranges overlapDeepSeek V4.1 Flash vs Pareto 26.9, Trails, ranges overlapDeepSeek V4.1 Flash vs Haiku 5.5 max, Trails, ranges overlapDeepSeek V4.1 Flash vs GPT 6 Sol medium, Trails, ranges overlapDeepSeek V4.1 Flash vs GPT 6 Luna max, Trails, ranges overlapDeepSeek V4.1 Flash vs Haiku 5.5 low, Leads, ranges overlapDeepSeek V4.1 Flash vs Opus 4.8, Leads, ranges overlapDeepSeek V4.1 Flash vs Muse control, Leads, ranges overlapDeepSeek V4.1 Flash vs GLM 5.3 Flash, Leads, ranges overlapDeepSeek V4.1 Flash vs GPT 6 Luna low, Leads, 95% ranges apart
10Haiku 5.5 lowHaiku 5.5 low vs SWE-2 Max, Trails, 95% ranges apartHaiku 5.5 low vs GPT 6.1 Sol medium, Trails, 95% ranges apartHaiku 5.5 low vs Opus 5, Trails, 95% ranges apartHaiku 5.5 low vs Grok 4.7 High, Trails, 95% ranges apartHaiku 5.5 low vs Pareto 26.9, Trails, 95% ranges apartHaiku 5.5 low vs Haiku 5.5 max, Trails, ranges overlapHaiku 5.5 low vs GPT 6 Sol medium, Trails, 95% ranges apartHaiku 5.5 low vs GPT 6 Luna max, Trails, ranges overlapHaiku 5.5 low vs DeepSeek V4.1 Flash, Trails, ranges overlapHaiku 5.5 low vs Opus 4.8, Leads, ranges overlapHaiku 5.5 low vs Muse control, Leads, ranges overlapHaiku 5.5 low vs GLM 5.3 Flash, Leads, ranges overlapHaiku 5.5 low vs GPT 6 Luna low, Leads, 95% ranges apart
11Opus 4.8Opus 4.8 vs SWE-2 Max, Trails, 95% ranges apartOpus 4.8 vs GPT 6.1 Sol medium, Trails, 95% ranges apartOpus 4.8 vs Opus 5, Trails, 95% ranges apartOpus 4.8 vs Grok 4.7 High, Trails, 95% ranges apartOpus 4.8 vs Pareto 26.9, Trails, ranges overlapOpus 4.8 vs Haiku 5.5 max, Trails, ranges overlapOpus 4.8 vs GPT 6 Sol medium, Trails, ranges overlapOpus 4.8 vs GPT 6 Luna max, Trails, ranges overlapOpus 4.8 vs DeepSeek V4.1 Flash, Trails, ranges overlapOpus 4.8 vs Haiku 5.5 low, Trails, ranges overlapOpus 4.8 vs Muse control, Leads, ranges overlapOpus 4.8 vs GLM 5.3 Flash, Leads, ranges overlapOpus 4.8 vs GPT 6 Luna low, Leads, ranges overlap
12Muse controlMuse control vs SWE-2 Max, Trails, 95% ranges apartMuse control vs GPT 6.1 Sol medium, Trails, 95% ranges apartMuse control vs Opus 5, Trails, 95% ranges apartMuse control vs Grok 4.7 High, Trails, 95% ranges apartMuse control vs Pareto 26.9, Trails, 95% ranges apartMuse control vs Haiku 5.5 max, Trails, 95% ranges apartMuse control vs GPT 6 Sol medium, Trails, 95% ranges apartMuse control vs GPT 6 Luna max, Trails, ranges overlapMuse control vs DeepSeek V4.1 Flash, Trails, ranges overlapMuse control vs Haiku 5.5 low, Trails, ranges overlapMuse control vs Opus 4.8, Trails, ranges overlapMuse control vs GLM 5.3 Flash, Leads, ranges overlapMuse control vs GPT 6 Luna low, Leads, ranges overlap
13GLM 5.3 FlashGLM 5.3 Flash vs SWE-2 Max, Trails, 95% ranges apartGLM 5.3 Flash vs GPT 6.1 Sol medium, Trails, 95% ranges apartGLM 5.3 Flash vs Opus 5, Trails, 95% ranges apartGLM 5.3 Flash vs Grok 4.7 High, Trails, 95% ranges apartGLM 5.3 Flash vs Pareto 26.9, Trails, 95% ranges apartGLM 5.3 Flash vs Haiku 5.5 max, Trails, 95% ranges apartGLM 5.3 Flash vs GPT 6 Sol medium, Trails, 95% ranges apartGLM 5.3 Flash vs GPT 6 Luna max, Trails, ranges overlapGLM 5.3 Flash vs DeepSeek V4.1 Flash, Trails, ranges overlapGLM 5.3 Flash vs Haiku 5.5 low, Trails, ranges overlapGLM 5.3 Flash vs Opus 4.8, Trails, ranges overlapGLM 5.3 Flash vs Muse control, Trails, ranges overlapGLM 5.3 Flash vs GPT 6 Luna low, Leads, ranges overlap
14GPT 6 Luna lowGPT 6 Luna low vs SWE-2 Max, Trails, 95% ranges apartGPT 6 Luna low vs GPT 6.1 Sol medium, Trails, 95% ranges apartGPT 6 Luna low vs Opus 5, Trails, 95% ranges apartGPT 6 Luna low vs Grok 4.7 High, Trails, 95% ranges apartGPT 6 Luna low vs Pareto 26.9, Trails, 95% ranges apartGPT 6 Luna low vs Haiku 5.5 max, Trails, 95% ranges apartGPT 6 Luna low vs GPT 6 Sol medium, Trails, 95% ranges apartGPT 6 Luna low vs GPT 6 Luna max, Trails, 95% ranges apartGPT 6 Luna low vs DeepSeek V4.1 Flash, Trails, 95% ranges apartGPT 6 Luna low vs Haiku 5.5 low, Trails, 95% ranges apartGPT 6 Luna low vs Opus 4.8, Trails, ranges overlapGPT 6 Luna low vs Muse control, Trails, ranges overlapGPT 6 Luna low vs GLM 5.3 Flash, Trails, ranges overlap
  • Leads, 95% ranges apart
  • Leads, ranges overlap
  • Tie
  • Trails, ranges overlap
  • Trails, 95% ranges apart
  • No Stet score

1SWE-2 Max

2GPT 6.1 Sol medium

3Opus 5

4Grok 4.7 High

5Pareto 26.9

6Haiku 5.5 max

7GPT 6 Sol medium

8GPT 6 Luna max

9DeepSeek V4.1 Flash

10Haiku 5.5 low

11Opus 4.8

12Muse control

13GLM 5.3 Flash

14GPT 6 Luna low