SEA-HELM (English) - automationbench: leaderboard

Metric: Normalized Score. Source: leaderboard.sea-lion.ai. Saturation forecast: Around February 2027. 61 models tracked. Updated 22 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GLM-5.3Reasoning49.34#79
2Qwen3.8-Flash-NextReasoning48.76#122
3Qwen-SEA-LION-v5-27BReasoning38.61
4Qwen 3.8 27BReasoning38.1#232
5DeepSeek V4 Flash (0731)Reasoning37.47#159
6DeepSeek V4 FlashReasoning14.87#273
7Qwen 3.6 27BReasoning14.66#358
8Gemma 4 26B A4B (IT)Reasoning14.22#563
9Gemma 4 31B (IT)Reasoning14#354
10Muse Glimmer 30BReasoning13.66#395
11Qwen-SEA-LION-v5-397B-A17B-PlusReasoning13.27#178
12Mistral-Medium-3.5-128BReasoning11.46
13Qwen-SEA-LION-v4.5-27B-ITReasoning10.49#538
14Qwen 3.6 35B A3BReasoning9.73#485
15Gemma 4 12B (IT)Reasoning9.19#732

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

Interactive version: theaggregate.ai/benchmark?slug=sea-helm-english-automationbench · How It Works · Data refreshed daily, snapshot 2026-11-22.