HLE (with tools)

HLE (with tools) - Humanity's Last Exam with tool use variant.

Category

STEM & Reasoning

Max Score

100.0

Score Type

percent

Active

Yes

Model Scores

Scores for HLE (with tools)

Model Score Date Verified
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16
42.58%
04.06.2026 Verified
GLM-5.1
70.13%
04.06.2026 Unverified
Kimi K2.6
77.75%
04.06.2026 Unverified
Qwen3.5-397B-A17B
65.68%
04.06.2026 Unverified
DeepSeek-V4-Pro
65.47%
04.06.2026 Unverified
DeepSeek-V4-Flash
58.90%
26.06.2026 Verified
Fable-5
100.00%
18.08.2026 Unverified
GPT-5.6 Sol
86.23%
18.08.2026 Unverified
Claude Opus 4.8
86.02%
18.08.2026 Unverified
Qwen3.8-2.4T-A95B
82.42%
18.08.2026 Verified
Qwen3.7-Max
76.69%
17.06.2026 Unverified
Kimi K3
81.99%
18.08.2026 Verified
GLM-5.2
79.24%
17.06.2026 Verified
Gemini 3.1 Pro
72.25%
17.06.2026 Unverified
GPT-5.5
73.94%
17.06.2026 Unverified
Qwen3.5-122B-A10B
63.98%
01.02.2026 Verified
Qwen3.5-27B
66.10%
01.02.2026 Unverified
Qwen3.5-35B-A3B
63.77%
01.02.2026 Unverified
GPT-5-mini 2025-08-07
39.19%
01.02.2026 Unverified
Ornith-1.5-35B-A3B
34.11%
19.08.2026 Verified
Ornith-1.0-35B-A3B
27.12%
19.08.2026 Unverified
Qwen3.6-35B-A3B
24.58%
19.08.2026 Unverified
Gemma-4-31B
19.49%
19.08.2026 Unverified
GPT-5.4
73.73%
23.08.2026 Verified
Claude Opus 4.6
75.64%
23.08.2026 Verified
Kimi K2.5
69.70%
23.08.2026 Verified
GPT-OSS-120B
3.60%
08.08.2025 Verified
GPT-OSS-20B
17.30
05.08.2025 Verified
DeepSeek-V4-Pro
58.05%
26.06.2026 Verified
DeepSeek-V4-Pro-0813
90.47%
13.08.2026 Verified
DeepSeek-V4-Flash-0731
72.46%
13.08.2026 Verified
Opus-4.8
86.02%
13.08.2026 Verified
Opus-4.6
75.85%
26.06.2026 Verified
Qwen3.8-Flash-Next
39.41%
26.08.2026 Verified
Qwen3.8-27B
28.60%
26.08.2026 Unverified
Qwen3.7-Plus
36.86%
26.08.2026 Unverified
Claude-Opus-4.6 (Max)
48.09%
26.08.2026 Unverified
GLM-5.3-Flash
80.51%
27.08.2026 Verified
DeepSeek-V4-Vision-Exp
80.08%
27.08.2026 Verified
GLM-5.3
95.76%
28.08.2026 Verified
Hy3
73.31%
28.08.2026 Verified
Hy4 preview
80.72%
28.08.2026 Verified
DeepSeek V4 Pro
90.47%
28.08.2026 Verified
Claude Opus 5
92.37%
28.08.2026 Verified
DeepSeek-V4.1-Flash
98.73%
10.09.2026 Verified