Can Ling 3.0 flash run on the 2x RTX 5090?
Yes. At 8K context the best fit is Q2_K, using about 43.8 GB of the 60.8 GB usable on the 2x RTX 5090. Estimated speed ceiling: ~888 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) • | 43.6 GB Fits | 43.8 GB Fits | 44.9 GB Fits | 49.4 GB Fits | 888 |
| 3-bit (Q3_K_M) • | 55.8 GB Tight | 56.0 GB Tight | 57.1 GB Tight | 61.5 GB No | 691 |
| 4-bit (Q4_K_M) • | 73.3 GB No | 73.5 GB No | 74.6 GB No | 79.1 GB No | — |
| 5-bit (Q5_K_M) • | 85.3 GB No | 85.5 GB No | 86.6 GB No | 91.0 GB No | — |
| 6-bit (Q6_K) • | 103 GB No | 103 GB No | 104 GB No | 108 GB No | — |
| 8-bit (Q8_0) • | 127 GB No | 127 GB No | 128 GB No | 133 GB No | — |
| 16-bit (F16/BF16) • | 238 GB No | 239 GB No | 240 GB No | 244 GB No | — |
• real GGUF file size from bartowski/Ling-3.0-flash-GGUF. Speed is a bandwidth ceiling (1792 GB/s), not a benchmark. Methodology.
More
Ling 3.0 flash on every GPU and Mac · Every model that runs on the 2x RTX 5090 · Try other settings in the calculator
Data: Hugging Face Hub (inclusionAI/Ling-3.0-flash), NVIDIA specs. Data as of 2026-11-17.