Still the benchmark for consumer AI. 24 GB VRAM fits DeepSeek R1 32B and Llama 4 Scout fully. 165 t/s on Llama 3.1 8B.
| VRAM | 24 GB |
| Memory Bandwidth | 1008 GB/s |
| TDP | 450 W |
| Architecture | Ada Lovelace AD102 |
| Release Year | 2022 |
| MSRP at Launch | $1,599 |
| Inference Speed (Llama 3.1 8B Q4_K_M) | ~165 tokens/sec |
| Inference Speed (Llama 3.3 70B Q4_K_M) | ~28 tokens/sec |
All models below run comfortably in 24 GB VRAM with Q4_K_M quantization.
| Llama 4 | 67 GB VRAM (smallest variant — needs more VRAM or a lower quant) · Q4_K_M · ollama run llama4:scout |
| Llama 3.3 | 43 GB VRAM (smallest variant — needs more VRAM or a lower quant) · 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 | 20 GB VRAM · Q4_K_M · ollama run qwen3:32b |
| Qwen 3.6 | 22 GB VRAM · Q4_K_M · ollama run qwen3.6:35b-a3b |
| Qwen 3.7 | 22 GB VRAM · Q4_K_M · qwen3-7 |
| Gemma 3 | 17 GB VRAM · Q4_K_M · ollama run gemma3:27b |
Install Ollama then run the recommended model for this GPU:
ollama run deepseek-r1:32b
Yes — the NVIDIA GeForce RTX 4090 has 24 GB VRAM and runs Still the benchmark for consumer AI. 24 GB VRAM fits DeepSeek R1 32B and Llama 4 Scout fully. 165 t/s on Llama 3.1 8B.
The NVIDIA GeForce RTX 4090 runs Llama 3.1 8B at ~165 tokens/sec with Q4_K_M quantization. For the 70B model it achieves ~28 tokens/sec.
With 24 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 deepseek-r1:32b.
← All GPU Reviews | Check Your Hardware | Full Benchmarks | Can I Run It?