Can Meta Llama Guard 2 8B run on the H200?
Yes. At 8K context the best fit is F16, using about 16.6 GB of the 134 GB usable on the H200. Estimated speed ceiling: ~299 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | Max tok/s (est., 8K) |
|---|---|---|---|
| 2-bit (Q2_K) • | 4.1 GB Fits | 4.6 GB Fits | 1510 |
| 3-bit (Q3_K_M) • | 4.9 GB Fits | 5.4 GB Fits | 1194 |
| 4-bit (Q4_K_M) • | 5.7 GB Fits | 6.2 GB Fits | 975 |
| 5-bit (Q5_K_M) • | 6.5 GB Fits | 7.0 GB Fits | 837 |
| 6-bit (Q6_K) • | 7.3 GB Fits | 7.8 GB Fits | 728 |
| 8-bit (Q8_0) • | 9.1 GB Fits | 9.6 GB Fits | 562 |
| 16-bit (F16/BF16) | 16.1 GB Fits | 16.6 GB Fits | 299 |
• real GGUF file size from tybrs/llama-guard-quant. Speed is a bandwidth ceiling (4800 GB/s), not a benchmark. Methodology.
What fits instead
- Cheapest listed GPU that runs Meta Llama Guard 2 8B: Arc B580 (12 GB, launch MSRP $249) at Q8_0, fully in VRAM at 8K context.
- Largest popular model that fits the H200: Qwen3.8 Flash Next (180B, MoE) at Q5_K_M, about 121 GB at 8K context.
"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.
More
Meta Llama Guard 2 8B on every GPU and Mac · Every model that runs on the H200 · Other models that fit 64 GB VRAM · Try other settings in the calculator
Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs
Data: Hugging Face Hub (meta-llama/Meta-Llama-Guard-2-8B), NVIDIA specs. Model data last checked 2026-11-17; refreshed hourly.