Can Llama 3.3 70B Instruct run on the 2x RTX 3090?
Yes. At 8K context the best fit is Q3_K_M, using about 35.3 GB of the 45.6 GB usable on the 2x RTX 3090. Estimated speed ceiling: ~27 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) | 26.5 GB Fits | 27.8 GB Fits | 35.3 GB Fits | 65.3 GB No | 35 |
| 3-bit (Q3_K_M) | 34.0 GB Fits | 35.3 GB Fits | 42.8 GB Tight | 72.8 GB No | 27 |
| 4-bit (Q4_K_M) | 42.1 GB Tight | 43.3 GB Tight | 50.8 GB No | 80.8 GB No | 22 |
| 5-bit (Q5_K_M) | 48.7 GB No | 50.0 GB No | 57.5 GB No | 87.5 GB No | — |
| 6-bit (Q6_K) | 55.8 GB No | 57.0 GB No | 64.5 GB No | 94.5 GB No | — |
| 8-bit (Q8_0) | 71.7 GB No | 73.0 GB No | 80.5 GB No | 110 GB No | — |
| 16-bit (F16/BF16) | 133 GB No | 135 GB No | 142 GB No | 172 GB No | — |
• real GGUF file size. Speed is a bandwidth ceiling (936 GB/s), not a benchmark. Methodology.
More
Llama 3.3 70B Instruct on every GPU and Mac · Every model that runs on the 2x RTX 3090 · Try other settings in the calculator
Data: Hugging Face Hub (meta-llama/Llama-3.3-70B-Instruct), NVIDIA specs. Data as of 2026-11-17.