LMArcade - terminus-2: leaderboard

LMArcade head-to-head human-voted playable-game Elo leaderboard inside the terminus-2 coding-agent harness (builds >= 5).

Metric: Elo Rating. Source: www.lmarcade.com. Saturation forecast: Around September 2027. 26 models tracked. Updated 22 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 51689#26
2Claude Fable 51687#32
3GPT-5.5Medium1686#56
4Claude Opus 4.71665#86
5Claude Sonnet 4.61656#179
6GPT-5.4Medium1605#138
7Claude Opus 4.81596#64
8Kimi K31560#62
9GLM-5.11542#200
10Qwen 3.6 Max1511#192
11Kimi K2.61497#197
12MiniMax-M2.71495#539
13Gemini 3.1 Pro (Preview)1491#102
14MiMo-V2.5-Pro1467#292
15MiniMax-M31460#290

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

Interactive version: theaggregate.ai/benchmark?slug=lmarcade-terminus-2 · How It Works · Data refreshed daily, snapshot 2026-11-22.