LMArcade - swe-agent: leaderboard

LMArcade head-to-head human-voted playable-game Elo leaderboard inside the swe-agent coding-agent harness (builds >= 5).

Metric: Elo Rating. Source: www.lmarcade.com. Saturation forecast: Around September 2027. 24 models tracked. Updated 22 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Fable 51720#32
2Claude Opus 51675#26
3Claude Opus 4.81639#64
4GPT-5.5Medium1607#56
5Claude Sonnet 51600#124
6Claude Opus 4.71596#86
7GPT-5.4Medium1582#138
8Claude Sonnet 4.61563#179
9MiMo-V2.5-Pro1554#292
10Gemini 3.1 Pro (Preview)1546#102
11Gemini 3 Flash1531#180
12Kimi K31526#62
13Qwen 3.6 Max1525#192
14DeepSeek V4 Pro1475#201
15GLM-5.11472#200

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=lmarcade-swe-agent · How It Works · Data refreshed daily, snapshot 2026-11-22.