Llama-Guard-4-12B
Llama-Guard-4-12B is an open-weight vision and language model from Meta with 12B parameters. At 4-bit it needs about 8.6 GB of memory, so the smallest current device that fits it is the Mac mini M6 (16GB).
Overview
Includes room for an 8k context and runtime overhead. More context needs more memory.
Key information
- Lab
- Meta
- Released
- 23 Apr 2025
- Parameters
- 12B
- Type
- Vision and language
- Context
- Unknown
- License
- Llama 4 license
- HF downloads, 30 days
- 130,983
- Repository
- meta-llama/Llama-Guard-4-12B
More from Meta
View allWill it run on your hardware?
How we estimate| Estimate or device | Memory |
|---|---|
| Q4 | 8.6 GB |
| Q8 | 15.1 GB |
| BF16 | 28.5 GB |
| RTX 5060 Ti 16 GB | 16 GB usable |
| RTX 5090 32 GB | 32 GB usable |
| DGX Spark 128 GB | 108.8 GB usable |
| Mac Studio M5 Ultra 256 GB | 192 GB usable |
Single-device memory estimate. Pick a device to update the On your device column.
36 GB usable memory
| Format | Memory needed | On your device | Example hardware |
|---|---|---|---|
| 4-bit (Q4) | ≈ 8.6 GB | Fits 24% | Mac mini M6 (16GB)16 GB memory |
| 8-bit (Q8) | ≈ 15.1 GB | Fits 42% | Arc Pro B6024 GB memory |
| 16-bit (BF16) | ≈ 28.5 GB | Fits 79% | Radeon AI PRO R970032 GB memory |
Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.
Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.
Where to run it
Through an API
DeepInfra (bf16) is the cheapest provider we found.
$0.144 / $0.144
Input / output per 1M tokens
₹14 / ₹14 at ₹94.29 per $1
Blended: $0.144 (₹14) per 1M. Ranked by a 3:1 mix of input and output tokens.
| Precision | |||||
|---|---|---|---|---|---|
| DeepInfraCheapest | $0.144₹14 | $0.144₹14 | $0.144₹14 | 163,840 | BF16 |
INR at ₹94.29 per $1. Prices fetched 14 Nov 2026.
More about this model
Frequently asked questions
How much memory does Llama-Guard-4-12B need? At an 8k context, about 8.6 GB at 4-bit, 15.1 GB at 8-bit and 28.5 GB at 16-bit (BF16). A longer context needs more.
Can I run Llama-Guard-4-12B on a Mac? The smallest current Mac that fits the 4-bit estimate of 8.6 GB is the Mac mini M6 (16GB). We have not found a trusted MLX or GGUF build yet.
Does the full context fit? We do not have a context length for this model. Our memory estimates use 8k, so long prompts need more memory than shown.
What does it cost through an API? The cheapest provider we found is DeepInfra, at $0.14 per 1M input tokens and $0.14 per 1M output tokens (₹14 and ₹14 at 94.29 rupees per dollar). Ranked by a 3:1 mix of input and output tokens.
License in plain English
Llama 4 license. Meta's community license for Llama. It allows commercial use with conditions, including an acceptable use policy and extra terms for very large services. This summary is not legal advice.
OMHQ RigBench
Not yet tested. We show no speed or quality score without a measurement.
Revision history and sources
Last updated . Sources:Hugging Face model card, OpenRouter.