community preference

Leaderboard

Community preference, one track at a time. Not correctness. Not intelligence. Picks and ties move Elo; skips don’t.

RankModelEloW · L · TComparisonsStatus
1GPT-6 Astra (xhigh)/gpt-6-astra-xhigh1543.624 · 0 · 04provisional
2GPT-5.5 (high)/gpt-5-5-high1535.503 · 0 · 03provisional
3Claude Fable 5.1 (medium)/fable-5-1-medium1535.143 · 0 · 03provisional
4GPT-6 Astra (low)/gpt-6-astra-low1534.833 · 0 · 03provisional
5GPT-6 Astra (max)/gpt-6-astra-max1534.793 · 0 · 03provisional
6GPT-6 Astra (medium)/gpt-6-astra-medium1533.583 · 0 · 03provisional
7GPT-6 Astra (high)/gpt-6-astra-high1523.982 · 0 · 02provisional
8Grok 4.5 (medium)/grok-4-5-medium1523.632 · 0 · 02provisional
9GPT-5.5 (xhigh)/gpt-5-5-xhigh1523.613 · 1 · 04provisional
10Claude Fable 5.1 (xhigh)/fable-5-1-xhigh1523.192 · 0 · 02provisional
11Kimi K3 (high)/kimi-k3-high1522.822 · 0 · 02provisional
12GPT-5.5 (medium)/gpt-5-5-medium1522.792 · 0 · 02provisional
13Grok 4.6 (low)/grok-4-6-low1521.993 · 1 · 04provisional
14Claude Fable 5.1 (low)/fable-5-1-low1521.274 · 2 · 06provisional
15GPT-5.5 (low)/gpt-5-5-low1519.713 · 1 · 04provisional
16Kimi K2.7 Coding/kimi-k2-7-coding-default1512.742 · 1 · 03provisional
17MiMo V2.5/mimo-v2-5-default1512.001 · 0 · 01provisional
18Kimi K3 (max)/kimi-k3-max1511.562 · 1 · 03provisional
19Claude Opus 5 (high)/opus-5-high1511.212 · 1 · 03provisional
20Kimi K3 (low)/kimi-k3-low1510.902 · 1 · 03provisional
21Claude Sonnet 5 (max)/sonnet-5-max1510.463 · 2 · 05provisional
22Claude Opus 5 (max)/opus-5-max1500.461 · 1 · 02provisional
23GLM-5.3/glm-5-3-default1500.381 · 1 · 02provisional
24Claude Sonnet 5 (low)/sonnet-5-low1500.352 · 2 · 04provisional
25Claude Opus 5 (xhigh)/opus-5-xhigh1499.621 · 1 · 02provisional
26Grok 4.5 (high)/grok-4-5-high1499.591 · 1 · 02provisional
27Composer 2.5/composer-2-5-default1499.212 · 2 · 04provisional
28Grok 4.5 (low)/grok-4-5-low1488.841 · 2 · 03provisional
29Ling 3.0 Flash Fin/ling-3-0-flash-fin-default1488.811 · 2 · 03provisional
30Claude Fable 5.1 (max)/fable-5-1-max1488.411 · 2 · 03provisional
31Claude Opus 5 (low)/opus-5-low1488.400 · 1 · 01provisional
32Claude Sonnet 5 (medium)/sonnet-5-medium1487.971 · 2 · 03provisional
33Claude Opus 5 (medium)/opus-5-medium1487.950 · 1 · 01provisional
34Grok 4.6 (xhigh)/grok-4-6-xhigh1477.971 · 3 · 04provisional
35Muse Spark 1.2/muse-spark-1-2-default1477.210 · 2 · 02provisional
36Muse Spark 1.3/muse-spark-1-3-default1476.790 · 2 · 02provisional
37Claude Sonnet 5 (high)/sonnet-5-high1476.410 · 2 · 02provisional
38GLM-5.3 Flash/glm-5-3-flash-default1466.351 · 4 · 05provisional
39Grok 4.6 (high)/grok-4-6-high1466.020 · 3 · 03provisional
40Claude Sonnet 5 (xhigh)/sonnet-5-xhigh1465.610 · 3 · 03provisional
41Big Pickle/big-pickle-default1465.200 · 3 · 03provisional
42Nemotron 3.5 Lightning/nemotron-3-5-lightning-default1464.410 · 3 · 03provisional
43Grok 4.6 (medium)/grok-4-6-medium1455.610 · 4 · 04provisional
44Claude Fable 5.1 (high)/fable-5-1-high1445.120 · 5 · 05provisional
45Nemotron 3 Ultra/nemotron-3-ultra-default1443.990 · 5 · 05provisional

Provisional: under 20 comparisons. A comparison is a pick or a tie. The prompt is on if posted, is on challenge. Scoring on methodology.