Can Apertus 8B Instruct 2509 run on the RTX 3080 10GB?
Yes. At 8K context the best fit is Q6_K, using about 7.8 GB of the 9.5 GB usable on the RTX 3080 10GB. Estimated speed ceiling: ~115 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|
| 2-bit (Q2_K) • | 4.2 GB Fits | 4.7 GB Fits | 7.7 GB Fits | 231 |
| 3-bit (Q3_K_M) • | 5.0 GB Fits | 5.5 GB Fits | 8.5 GB Fits | 182 |
| 4-bit (Q4_K_M) • | 5.9 GB Fits | 6.4 GB Fits | 9.4 GB Tight | 150 |
| 5-bit (Q5_K_M) • | 6.6 GB Fits | 7.1 GB Fits | 10.1 GB No | 131 |
| 6-bit (Q6_K) • | 7.3 GB Fits | 7.8 GB Fits | 10.8 GB No | 115 |
| 8-bit (Q8_0) • | 9.1 GB Tight | 9.6 GB No | 12.6 GB No | — |
| 16-bit (F16/BF16) • | 16.2 GB No | 16.7 GB No | 19.7 GB No | — |
• real GGUF file size from unsloth/Apertus-8B-Instruct-2509-GGUF. Speed is a bandwidth ceiling (760 GB/s), not a benchmark. Methodology.
More
Apertus 8B Instruct 2509 on every GPU and Mac · Every model that runs on the RTX 3080 10GB · Other models that fit 12 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (swiss-ai/Apertus-8B-Instruct-2509), NVIDIA specs. Data as of 2026-11-17.