Benchgen

TerminalBench 2.1 — Results

RankModelScore
1gpt-5-6-sol88.8
2kimi-k388.3
3qwen3-8-max86.6
4gpt-5-583.4
5muse-spark-1-282.9
6fugu-ultra82.1
7claude-sonnet-580.4
8fugu80.2
9claude-opus-4-771.7
10seed-2-1-pro71
11gemini-3-1-pro70.7
12laguna-s-2-170.2
13seed-2-1-turbo67.6
14claude-sonnet-4-667
15inkling63.8
16solar-pro-457
17muse-glimmer51.7
18kimi-k2-thinking-090547.1
19minimax-m246.3
20nova-2-pro41.3
21deepseek-v3-2-exp37.7
22glm-4-733.3
23deepseek-v3-131.3
24kimi-k2-instruct30
25nemotron-3-super-120b-a12b25.8
T

TerminalBench 2.1 — Phases

1 phaseActive

Phase 1·

TerminalBench 2.1

Jun 22, 2026
Current