IBM · granite
Can I run granite4.1-guardian?
Granite Guardian 4.1 is a specialized safety and judging model from IBM Research that evaluates whether LLM prompts and responses meet specified harm criteria.
8.4B parameters128K contexttextApache License5 builds
granite4.1-guardian system requirements
Memory needed at 4K context, counting model weights, KV cache, compute buffers and runtime overhead. Download sizes come from the official Ollama registry.
| Build | Quantization | Download | Memory needed |
|---|---|---|---|
| granite4.1-guardian:8b-q4_0 | Q4_0 | 4.5 GB | ~6.8 GB(est.) |
| granite4.1-guardian:8b-q4_K_M | Q4_K_M | 4.8 GB | ~7.2 GB(est.) |
| granite4.1-guardian:8b-q5_K_M | Q5_K_M | 5.6 GB | ~8.3 GB(est.) |
| granite4.1-guardian:8b | Q6_K | 6.4 GB | ~9.6 GB(est.) |
| granite4.1-guardian:8b-q8_0 | Q8_0 | 8.3 GB | ~12.3 GB(est.) |
Figures marked (est.) approximate the KV cache because this model's architecture details are not published. They are indicative rather than exact.
How to run granite4.1-guardian locally
- 1.Install Ollama for Windows, macOS or Linux.
- 2.Run this in a terminal:
ollama run granite4.1-guardian:8b-q4_0 - 3.The weights download on first run, then an interactive prompt opens.
Frequently asked questions about granite4.1-guardian
- How much VRAM do I need to run granite4.1-guardian?
- At 4K context, the smallest published build of granite4.1-guardian (Q4_0) needs roughly 6.8 GB of GPU memory once weights, KV cache, compute buffers and runtime overhead are counted. Bigger quantizations and longer context windows need more. It can also run partly or entirely on the CPU using system RAM, more slowly.
- Can I run granite4.1-guardian without a dedicated GPU?
- Yes, but slowly. Without a GPU the model runs on the CPU using system RAM, which usually means a few tokens per second rather than dozens. You would need at least 6.8 GB of free RAM for the smallest build.
- How big is the granite4.1-guardian download?
- The smallest published build is 4.5 GB. There are 5 builds in total, the largest being 8.3 GB. Leave some extra free disk space beyond the download itself.
- How do I run granite4.1-guardian locally?
- Install Ollama, then run "ollama run granite4.1-guardian:8b-q4_0" in a terminal. The weights download on first use and an interactive session opens.
Will granite4.1-guardian run on your PC?
Scan your hardware once and get a verdict for this model and every build — free, no account.
Model data from the official Ollama registry · verified 13/09/2026