List your company
ResearchFormerly WizWand

General Competency Assessment on General-Bench

85.1Average ScoreClaude-Haiku-4.5
49.563258.789168.01577.2409Jul 2, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Claude-Haiku-4.5Model Scale=>100B2026.0785.1
Qwen3.5-FlashModel Scale=>100B2026.0784.42
GPT-5.4-MiniModel Scale=>100B2026.0780.72
Gemini-3.1-FlashModel Scale=>100B2026.0780.25
Doubao-Seed-2-LiteModel Scale=>100B2026.0780.13
Deepseek-V3.2Model Scale=>100B2026.0778.02
Gemma-3-27BModel Scale=10B<Scale<100B2026.0775.71
FitOne-32BModel Scale=32B2026.0775.19
GPT-OSS-20BModel Scale=10B<Scale<100B2026.0774.76
GLM-4-32B-0414Model Scale=10B<Scale<100B2026.0773.49
Qwen3-30B-A3BModel Scale=10B<Scale<100B2026.0772.76
Qwen3-32BModel Scale=32B2026.0772.67
FitOne-8BModel Scale=8B2026.0770.83
Qwen3-8BModel Scale=8B2026.0767.9
GLM-4-9B-0414Model Scale=9B2026.0762.27
InternLM3-8BModel Scale=8B2026.0755.35
Llama-3.1-8BModel Scale=8B2026.0752.41
Ministral-3-8BModel Scale=8B2026.0750.93