Written by Jakub Rusinowski · Last updated February 10, 2026
Yes — comfortably
Yes, comfortably — EXAONE 3.5 7.8B at Q8_0 needs about 10.2 GB of the 16 GB usable on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM), leaving ~5.8 GB spare and running at ~69 tok/s (estimated), with room for about 32,768 tokens of context.
Confidence: high · Recommended quantization: Q8_0 · Estimated speed: ~69 tok/s
| Usable memory for models | 16 GB |
| Memory bandwidth | 960 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
| Quant | Memory needed | Fits 16 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 17.5 GB | ✗ No | — | — | 15.6 GB |
| Q8_0 | 10.2 GB | ✓ Yes | 32K | ~69 tok/s | 8.3 GB |
| Q6_K | 8.3 GB | ✓ Yes | 32K | ~84.2 tok/s | 6.4 GB |
| Q5_K_M | 7.4 GB | ✓ Yes | 32K | ~93.7 tok/s | 5.5 GB |
| Q4_K_M | 6.6 GB | ✓ Yes | 32K | ~104.9 tok/s | 4.7 GB |
| Q3_K_M | 5.2 GB | ✓ Yes | 32K | ~131.4 tok/s | 3.3 GB |
| Q2_K | 4.4 GB | ✓ Yes | 32K | ~152.6 tok/s | 2.6 GB |
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| EXAONE 3.5 32B | 22.3 GB | ✗ Too large | — |
| EXAONE 3.5 7.8B | 6.6 GB | ✓ Fits | ~104.9 tok/s |
| EXAONE 3.5 2.4B | 2.9 GB | ✓ Fits | ~213 tok/s |
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Yes, comfortably — EXAONE 3.5 7.8B at Q8_0 needs about 10.2 GB of the 16 GB usable on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM), leaving ~5.8 GB spare and running at ~69 tok/s (estimated), with room for about 32,768 tokens of context.
Q8_0 — it needs about 10.2 GB of the 16 GB available, downloads as roughly 8.3 GB, and runs at an estimated 69 tokens/sec with up to 32K of context.
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
← Can I Run It? | EXAONE 3.5 model page | Check your hardware