Meta Llama Guard 2 8B — Hardware Requirements & GPU Compatibility
Chat
Meta Llama Guard 2 8B is a 8.0B-parameter open language model from Meta in the Llama family. At BF16 it needs about 17.67 GB of VRAM — see which GPUs and Macs can run it below.
Runs well on a 24 GB GPU
Runs on 24 GB (BF16 · ~37 tok/s · 32K)Macs from 24 GB (BF16 · ~12 tok/s)
Llama 3 Community
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Meta Llama Guard 2 8B
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve meta-llama/Meta-Llama-Guard-2-8B --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX PRO 4500 (32 GB) (11 GB headroom) · BF16
- Generation speed
- ~33 tok/s
- generation speed
- Cost per 1M output tokens
- $2.86
- per 1M output tokens
How Much VRAM Does Meta Llama Guard 2 8B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 17.7 GB | 23.2 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 32K context · ~37 tok/s |
Selected BF16
Brain floating point 1620.2GBat 16K context
- Weights
- 17.0 GB
- KV cache
- 2.9 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 3.8 GB free · ~37 tok/s
≥50 tok/s needs ≥1.36 TB/s · RTX 5090 ~66
Which Hardware Can Run Meta Llama Guard 2 8B?
BF16 · 17.7 GBMeta Llama Guard 2 8B (BF16) requires 17.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 23+ GB is recommended. 8 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 122 devices with unified memory can run Meta Llama Guard 2 8B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M5 Pro (24 GB).
20 GB
tight · up to 8K ctx
1 GPU
RX 7900 XT~27Simulate output on RX 7900 XT32–40 GB
fits · 11.8 GB free · up to 32K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~66Simulate output on RTX 5090M5 Max36 GB~18Simulate output on Mac Studio M5 Max (36 GB)M532 GB~6Simulate output on MacBook Air 13" M5 (32 GB)Orin~8Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~5Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at BF16
Alternatives to Meta Llama Guard 2 8B
Llama family · 19 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| ALoDLM 8BAmazon | — | 7.6 GB | ~119 | same size |
| Llm Jp 4.1 8B Thinkingllm-jp | — | 7.6 GB | ~115 | same size |
| Ling 3.0 TinyInclusion AI | — | 5.5 GB | ~364 | same size |
| LFM2.5 8B A1BLiquid AI | — | 5.6 GB | ~335 | same size |
| BitCPM CANN 8BOpenBMB | — | 5.6 GB | ~127 | same size |
| Hy MT2 7BTencent | — | 7.3 GB | ~122 | same size |
Specifications
- Publisher
- Meta
- Family
- Llama
- Parameters
- 8.0B
- Release Date
- 2024-04-17
- License
- Llama 3 Community
Frequently Asked Questions
How much VRAM does Meta Llama Guard 2 8B need?
Meta Llama Guard 2 8B requires 17.7 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 8.0B × 16 bits ÷ 8 = 16.1 GB
KV Cache + Overhead ≈ 1.6 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
VRAM usage by quantization
Can I run Meta Llama Guard 2 8B on a Mac?
Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run Meta Llama Guard 2 8B. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 17.7 GB of usable unified memory (RAM minus macOS overhead).
Can I run Meta Llama Guard 2 8B locally?
Yes — Meta Llama Guard 2 8B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Meta Llama Guard 2 8B?
At BF16, Meta Llama Guard 2 8B can reach ~272 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~37 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 17.7 × 0.65 = ~294 tok/s
Estimated speed at BF16 (17.7 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Meta Llama Guard 2 8B?
At BF16, the download is about 16.06 GB.
Which GPUs can run Meta Llama Guard 2 8B?
8 consumer GPUs can run Meta Llama Guard 2 8B at BF16. Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XTX. 1 GPU has plenty of headroom for comfortable inference.
Which devices can run Meta Llama Guard 2 8B?
122 devices with unified memory can run Meta Llama Guard 2 8B at BF16 (17.7 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.