Can ERNIE 4.5 21B A3B Thinking run on the H100 PCIe?
Yes. At 8K context the best fit is F16, using about 41.7 GB of the 76.0 GB usable on the H100 PCIe. Estimated speed ceiling: ~46 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | 8.4 GB Fits | 8.6 GB Fits | 9.9 GB Fits | 15.2 GB Fits | 245 |
| 3-bit (Q3_K_M) • | 10.6 GB Fits | 10.8 GB Fits | 12.1 GB Fits | 17.4 GB Fits | 190 |
| 4-bit (Q4_K_M) • | 13.2 GB Fits | 13.5 GB Fits | 14.8 GB Fits | 20.0 GB Fits | 150 |
| 5-bit (Q5_K_M) • | 15.3 GB Fits | 15.5 GB Fits | 16.9 GB Fits | 22.1 GB Fits | 128 |
| 6-bit (Q6_K) • | 17.6 GB Fits | 17.8 GB Fits | 19.1 GB Fits | 24.3 GB Fits | 111 |
| 8-bit (Q8_0) • | 22.4 GB Fits | 22.6 GB Fits | 24.0 GB Fits | 29.2 GB Fits | 86 |
| 16-bit (F16/BF16) • | 41.5 GB Fits | 41.7 GB Fits | 43.0 GB Fits | 48.3 GB Fits | 46 |
• real GGUF file size from unsloth/ERNIE-4.5-21B-A3B-Thinking-GGUF. Speed is a bandwidth ceiling (2000 GB/s), not a benchmark. Methodology.
What fits instead
- Cheapest listed GPU that runs ERNIE 4.5 21B A3B Thinking: Arc A770 16GB (16 GB, launch MSRP $349) at Q4_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
ERNIE 4.5 21B A3B Thinking 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
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 (baidu/ERNIE-4.5-21B-A3B-Thinking), NVIDIA specs. Model data last checked 2026-11-17; refreshed hourly.