Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Factuality and Hallucination Detection on FactScore
Loading...
96.1
FactScore
GPT-5-mini
48.572
60.911
73.25
85.589
Jun 30, 2026
FactScore
Factuality Score
Updated 23d ago
Evaluation Results
Method
Method
Links
FactScore
Factuality Score
GPT-5-mini
Tier=High
2026.06
96.1
-
Gemini-3-pro High
2026.06
92.6
-
Gemini-3-Flash
Tier=High
2026.06
92
-
GPT-5.2 High
2026.06
91.9
-
Claude-Opus-4.5
2026.06
91.1
-
Claude-Sonnet-4.5
2026.06
90.6
-
Seed2.0 Pro
2026.06
71.2
-
Seed2.0 Lite
2026.06
62.4
-
Seed2.0 Mini
2026.06
50.4
-
Feedback
Search any
task
Search any
task