Can Nous Hermes 2 Yi 34B run on the H100 PCIe?
Yes. At 8K context the best fit is F16, using about 66.5 GB of the 76.0 GB usable on the H100 PCIe. Estimated speed ceiling: ~29 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | Max tok/s (est., 8K) |
|---|---|---|
| 2-bit (Q2_K) • | 13.9 GB Fits | 150 |
| 3-bit (Q3_K_M) • | 17.5 GB Fits | 116 |
| 4-bit (Q4_K_M) • | 21.2 GB Fits | 94 |
| 5-bit (Q5_K_M) • | 24.6 GB Fits | 80 |
| 6-bit (Q6_K) • | 28.2 GB Fits | 70 |
| 8-bit (Q8_0) | 35.5 GB Fits | 55 |
| 16-bit (F16/BF16) | 65.5 GB Fits | 29 |
• real GGUF file size from mradermacher/Nous-Hermes-2-Yi-34B-i1-GGUF. Speed is a bandwidth ceiling (2000 GB/s), not a benchmark. Methodology.
What fits instead
- Cheapest listed GPU that runs Nous Hermes 2 Yi 34B: RTX 5090 (32 GB, launch MSRP $1,999) at Q5_K_M, fully in VRAM at 8K context.
- Largest popular model that fits the H100 PCIe: gpt-oss 120B (117B, MoE) at F16, about 61.9 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
Nous Hermes 2 Yi 34B on every GPU and Mac · Every model that runs on the H100 PCIe · Other models that fit 64 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (NousResearch/Nous-Hermes-2-Yi-34B), NVIDIA specs. Model data last checked 2026-11-17; refreshed hourly.