Model ID
ibm-granite/granite-4.2-8b
Capabilities
Granite 4.2 8B
Granite 4.2 8B is a small-size reasoning model designed for coding, mathematics, tool calling, agentic workflows, and multilingual dialogue. It features 8B parameters with a 40-layer dense transformer using 4,096 hidden size, 32 attention heads, and 8 KV heads. The architecture employs Grouped Query Attention, RoPE with a 10M theta, and a 12,800-dimensional SwiGLU feed-forward network, with native reasoning and flexible thinking modes for balancing quality and latency. Supporting 128K native context with extension to 512K, it is optimized for efficient long-context reasoning and enterprise applications.
About this model
Granite 4.2 8B is a small-size reasoning model designed for coding, mathematics, tool calling, agentic workflows, and multilingual dialogue. It features 8B parameters with a 40-layer dense transformer using 4,096 hidden size, 32 attention heads, and 8 KV heads. The architecture employs Grouped Query Attention, RoPE with a 10M theta, and a 12,800-dimensional SwiGLU feed-forward network, with native reasoning and flexible thinking modes for balancing quality and latency. Supporting 128K native context with extension to 512K, it is optimized for efficient long-context reasoning and enterprise applications.
Model specifications and capabilities are published by the model author and reproduced here from HF Model Card (ibm-granite/granite-4.2-8b). Released under Apache 2.0. Further reading: Blog
Deploy Private Inference Endpoint
Choose your hardware and inference engine to get deployment commands and performance benchmarks tailored to your infrastructure
Hardware
B200
1,440 GB
Engine
vLLM
SGLang
Deployment code
Access Restricted
Enter your email to get access to this content
Ensure that granite_thinking_parser.py is present inside mapped local model directory prior to initialization to satisfy the container's plugin path requirement.
Performance under load
Tokens/s (total)
Benchmarks measured by Vultr on B200 · vLLM. Throughput and latency vary with concurrency, input length and engine configuration, so treat these as a comparison baseline, not a service guarantee.
NVIDIA HGX B200
Unlock Next-Generation AI Power on Vultr Cloud GPU