The reference datacenter GPU for LLM serving: 80 GB HBM3, FP8 transformer engine, first-class vLLM/TensorRT-LLM support. One H100 serves a 70B model to a whole department; 2–4× serve 100B+ MoE models. No official MSRP — figure shown is a typical single-unit PCIe street price; most businesses buy it inside an OEM server or rent it hourly (see the cloud directory) before committing.
| VRAM | 80 GB |
| Memory Bandwidth | 2000 GB/s |
| TDP | 350 W |
| Architecture | Hopper GH100 |
| Release Year | 2022 |
| MSRP at Launch | $25,000 |
| Inference Speed (Llama 3.1 8B Q4_K_M) | ~190 tokens/sec |
| Inference Speed (Llama 3.3 70B Q4_K_M) | ~40 tokens/sec |
All models below run comfortably in 80 GB VRAM with Q4_K_M quantization.
| Llama 4 | 67 GB VRAM · Q4_K_M · ollama run llama4:scout |
| Llama 3.3 | 43 GB VRAM · Q2_K_XS (Tight) · ollama run llama3.3 |
| Llama 3.1 Family | 6 GB VRAM · Q4_K_M · ollama run llama3.1 |
| DeepSeek R1 | 20 GB VRAM · Q4_K_M · ollama run deepseek-r1:32b |
| Qwen 3 | 80 GB VRAM · Q4_K_M · ollama run qwen3:235b-a22b |
| Qwen 3.5 | 74 GB VRAM · Q4_K_M · ollama run qwen3.5:122b |
| Qwen 3.6 | 22 GB VRAM · Q4_K_M · ollama run qwen3.6:35b-a3b |
| Gemma 3 | 17 GB VRAM · Q4_K_M · ollama run gemma3:27b |
Install Ollama then run the recommended model for this GPU:
ollama run llama3.3:70b
Yes — the NVIDIA H100 80GB has 80 GB VRAM and runs The reference datacenter GPU for LLM serving: 80 GB HBM3, FP8 transformer engine, first-class vLLM/TensorRT-LLM support.
The NVIDIA H100 80GB runs Llama 3.1 8B at ~190 tokens/sec with Q4_K_M quantization. For the 70B model it achieves ~40 tokens/sec.
With 80 GB you can run: Llama 4, Llama 3.3, Llama 3.1 Family, DeepSeek R1, Qwen 3. Use Ollama for the easiest setup: ollama run llama3.3:70b.
← All GPU Reviews | Check Your Hardware | Full Benchmarks | Can I Run It?