IBM's Granite 4.1 8B is a dense, decoder-only 8-billion-parameter instruction model built for enterprise text workflows, including tool calling, retrieval-augmented generation, code generation with fill-in-the-middle support, summarization, classification, extraction, and multilingual assistance.
Added Apr 29, 2026
Context Window
131.1K
Max Output
131.1K
Input Price (Auto)
$0.052/1M
Output Price (Auto)
$0.10/1M
Cache Read (Auto)
$0.052/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
12.4
Choose explicit providers for this model. Auto routing remains available as the default option.
Loading provider options…
Coding Index
7.3
Agentic Index
10.7
GPQA Diamond
Graduate-level scientific reasoning
43.3%
Better than 26% of models compared
HLE
Humanity's Last Exam
3.8%
Better than 11% of models compared
IFBench
Instruction-following benchmark
38.6%
Better than 42% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
27.8%
Better than 39% of models compared
AA-LCR
Long context reasoning evaluation
12.0%
Better than 28% of models compared
GDPval-AA
Economically valuable tasks
2.1%
Better than 44% of models compared
CritPt
Research-level physics reasoning
0.0%
Better than 36% of models compared
SciCode
Python programming for scientific computing
21.8%
Better than 28% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
0.0%
Better than 6% of models compared
AA-Omniscience Accuracy
Proportion of correctly answered questions
12.1%
Better than 20% of models compared
AA-Omniscience Hallucination Rate
Rate of incorrect answers among non-correct responses
87.3%
Better than 28% of models compared
Last updated May 15, 2026
Artificial Analysis
