JobBench

JobBench - in-house benchmark for evaluating professional job task completion.

Category

General Agents

Max Score

100.0

Score Type

percent

Active

Yes

Model Scores

Scores for JobBench

Model Score Date Verified
Qwen3.8-27B
25.11%
18.08.2026 Verified
Qwen3.6-27B
21.80
18.08.2026 Unverified
Qwen3.7-Plus
12.55%
18.08.2026 Unverified
Fable-5
77.06%
18.08.2026 Unverified
GPT-5.6 Sol
51.08%
18.08.2026 Unverified
Claude Opus 4.8
57.58%
18.08.2026 Unverified
Qwen3.7-Max
20.56%
18.08.2026 Unverified
Qwen3.8-2.4T-A95B
68.40%
18.08.2026 Verified
Kimi K3
70.35%
18.08.2026 Verified
Qwen3.8-Flash-Next
73.38%
26.08.2026 Verified
DeepSeek-V4-Flash-0731
42.21%
26.08.2026 Unverified
Hy3
27.71%
28.08.2026 Verified
Hy4 preview
86.36%
28.08.2026 Verified
DeepSeek V4 Pro
69.91%
28.08.2026 Verified
GLM-5.3
78.79%
28.08.2026 Verified
Claude Opus 5
100.00%
28.08.2026 Verified
Claude-Opus-4.6 (Max)
32.03%
26.08.2026 Unverified