← Back to Models

granite-4-h-tiny

ibm/granite-4-h-tiny

View on HuggingFace
Publisher
ibm
Architecture
granitehybrid
Parameters
7B
Quantization
Q4_K_M (4.0 bpw)
Format
gguf
Max Context
1049k tokens
Type
llm
File Size
3.94 GB
Est. RAM
5.4 GB
Est. VRAM
5.4 GB
Capabilities
trained_for_tool_use

Benchmark Results

Compare Models