IBM

Granite 3.3 8B Instruct

Multimodal
Zero-eval
#2AttaQ
#2TruthfulQA
#2AlpacaEval 2.0
+1 more

by IBM

+
+
+
+
About

Granite 3.3 8B Instruct was created as the instruction-tuned version of the Granite base model, designed to follow instructions reliably in enterprise settings. Built to serve business applications requiring dependable language understanding, it provides IBM's accessible interface for enterprise AI.

+
+
+
+
Timeline
AnnouncedApr 16, 2025
ReleasedApr 16, 2025
Knowledge CutoffApr 1, 2024
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Apache 2.0
Performance Overview
Performance metrics and category breakdown

Overall Performance

14 benchmarks
Average Score
69.8%
Best Score
89.7%
High Performers (80%+)
5
+
+
+
+
All Benchmark Results for Granite 3.3 8B Instruct
Complete list of benchmark scores with detailed information
HumanEval
text
0.90
89.7%
Self-reported
AttaQ
text
0.89
88.5%
Self-reported
HumanEval+
text
0.86
86.1%
Self-reported
AIME 2024
text
0.81
81.2%
Self-reported
GSM8k
text
0.81
80.9%
Self-reported
IFEval
text
0.75
74.8%
Self-reported
BIG-Bench Hard
text
0.69
69.1%
Self-reported
MATH-500
text
0.69
69.0%
Self-reported
TruthfulQA
text
0.67
66.9%
Self-reported
MMLU
text
0.66
65.5%
Self-reported
Showing 1 to 10 of 14 benchmarks
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+