Decision models · IBM

Granite Guardian 4.1 8B

Name: Granite Guardian 4.1 8B
Author: IBM

Task: Safety and policy checks
Returns: Fixed labels
Precision tier: 84 / 100
Size: 8.4B
Max input: 128K tokens
License: apache-2.0
Released: 2026-04

What it decides

Returns: Yes or no inside <score> tags, with optional step-by-step reasoning in think mode; also checks RAG groundedness and tool calls.
Checks: Harm, jailbreaks, RAG groundedness and relevance, tool-call hallucination, custom criteria
Languages: English
Precision: 84 / 100 · Out-of-distribution safety F1 0.79; RAG hallucination balanced accuracy 0.76 (model card).
Clarity: Fixed labels (12 of 15 clarity points)
How it runs: Writes a short answer · a small language model, so it runs in the same apps as LLMs.
License: apache-2.0 (Open (Apache, MIT))

Source: model card on Hugging Face (opens in a new tab) · How decision models are rated

How to run it

Quickest · Ollama

One command downloads a ready-to-run version and starts a chat.

Get Ollama at ollama.com/download (opens in a new tab) · model page (opens in a new tab)

ollama run granite4.1-guardian:8b

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/ibm-granite/granite-guardian-4.1-8b-GGUF (opens in a new tab)

ollama run hf.co/ibm-granite/granite-guardian-4.1-8b-GGUF

Original weights · Hugging Face

Full-precision files from IBM, for fine-tuning, vLLM, Transformers.

huggingface.co/ibm-granite/granite-guardian-4.1-8b (opens in a new tab)

No sign-in needed.

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

Apps that run Granite Guardian 4.1 8B

Our pick for each kind of computer, assuming the model fits in its memory. Pick your own GPU or Mac in the Model Finder for an exact answer.

Mac (Apple silicon)

Ollama

One command with Ollama’s own build of this model.

Also good: LM Studio, llama.cpp

Windows PC, NVIDIA GPU

Ollama

One command with Ollama’s own build of this model.

Also good: LM Studio, llama.cpp

Windows PC, AMD Radeon

Lemonade

AMD’s own app: llama.cpp tuned for Radeon on Vulkan and ROCm.

Also good: LM Studio, Ollama

Linux, NVIDIA GPU

Ollama

One command with Ollama’s own build of this model.

Also good: LM Studio, llama.cpp

Linux server, 80 GB+ GPUs

vLLM

The full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.

Also good: Ollama, LM Studio

App	Type	Opens it as	Systems	Engine	Start
LM Studio	Desktop app	GGUF	macOSWindowsLinux	llama.cpp, MLX	Open in LM Studio
Ollama	CLI + API	Ollama library	macOSWindowsLinux	llama.cpp, MLX	`ollama run granite4.1-guardian:8b`
llama.cpp	CLI + API	GGUF	macOSWindowsLinux	llama.cpp	`llama-server -hf ibm-granite/granite-guardian-4.1-8b-GGUF:Q4_K_M`
Jan	Desktop app	GGUF	macOSWindowsLinux	llama.cpp, MLX	Open in Jan
vLLM	Server	Original weights	WSLWindowsLinux	vLLM	`vllm serve ibm-granite/granite-guardian-4.1-8b`
SGLang	Server	Original weights	WSLWindowsLinux	SGLang	`python3 -m sglang.launch_server --model-path ibm-granite/granite-guardian-4.1-8b`
Docker Model Runner	CLI + API	GGUF	macOSWindowsLinux	llama.cpp, vLLM	`docker model run hf.co/ibm-granite/granite-guardian-4.1-8b-GGUF:Q4_K_M`
Lemonade	Desktop app	GGUF	macOSWindowsLinux	llama.cpp, ONNX Runtime GenAI	`lemonade pull ibm-granite/granite-guardian-4.1-8b-GGUF:Q4_K_M`
Unsloth Desktop	Desktop app	GGUF	macOSWindowsLinux	llama.cpp, MLX, stable-diffusion.cpp	Open in Unsloth Desktop
Atomic Chat	Desktop app	GGUF	macOSWindowsLinux	llama.cpp, MLX-VLM	Open in Atomic Chat
LocalAI	Server	GGUF	macOSWSLWindowsLinux	llama.cpp, rerankers, Transformers	Get LocalAI
Sentence Transformers	Library	Original weights	macOSWindowsLinux	PyTorch	Get Sentence Transformers

Which apps can open a model depends on the files published for it: GGUF for llama.cpp-based apps, MLX for Apple’s engine, the original weights for vLLM and SGLang. Compare every app.

Memory and speed

Each quantization on an NVIDIA GeForce RTX 4090, for one decision on a 512-token input. The best hardware table below covers other GPUs and Macs.

Format	Memory needed	Time per decision	Decisions per second	Quality	Download
`Q4_K_M`	6.0 GB	117 ms	8.6/s	79Great quality (A) · 79 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	6.9 GB	125 ms	8.0/s	81Great quality (A) · 81 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q6_K`	7.8 GB	135 ms	7.4/s	83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	9.8 GB	156 ms	6.4/s	84Great quality (A) · 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	17.7 GB	249 ms	4.0/s	84Great quality (A) · 84 of 100: how capable the model is at F16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab) Original weights (opens in a new tab)

One request at a time. Servers that batch (Text Embeddings Inference, vLLM, Infinity) handle many more decisions per second.

Best hardware for Granite Guardian 4.1 8B

The least expensive GPU in our catalog that runs it without offloading is the Intel Arc A580 (launch price $179) at Q4_K_M.

Quality	Hardware	Quant	Memory	Time per decision	Runs
84Great quality (A) · 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX PRO 6000 Blackwell	`Q8_0`	14.6 GB / 96.0 GB	97 ms	Runs great
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5090	`Q6_K`	9.4 GB / 32.0 GB	94 ms	Runs great
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking.	AMD Radeon RX 7900 XTX	`Q6_K`	9.0 GB / 24.0 GB	132 ms	Runs great
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4090	`Q6_K`	9.0 GB / 24.0 GB	135 ms	Runs well
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5080	`Q6_K`	8.6 GB / 16.0 GB	171 ms	Runs well
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking.	AMD Radeon RX 9070 XT	`Q6_K`	8.6 GB / 16.0 GB	182 ms	Runs well
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4080 SUPER	`Q6_K`	8.6 GB / 16.0 GB	198 ms	Runs well
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5070 Ti	`Q6_K`	8.6 GB / 16.0 GB	206 ms	Runs well
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4070 Ti SUPER	`Q6_K`	8.6 GB / 16.0 GB	227 ms	Runs well
83Great quality (A) · 83 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 3090	`Q6_K`	9.0 GB / 24.0 GB	234 ms	Runs well

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.