Can I Run EXAONE 3.5 2.4B on NVIDIA L40S?
Written by Jakub Rusinowski · Last updated July 21, 2026
Yes. EXAONE 3.5 2.4B needs 2.9 GB at 8K context (1.4 GB weights + 1.5 GB KV cache/overhead) at Q4_K_M and the NVIDIA L40S has 48 GB, leaving about 45.1 GB of headroom, at roughly 225 tok/s (estimated) with room for up to 32K of context.
See what else this hardware can run →
The numbers
| Model | EXAONE 3.5 2.4B |
| Parameters | 2.4B |
| Quantization | Q4_K_M |
| VRAM needed (8K context) | 2.9 GB |
| NVIDIA L40S VRAM | 48 GB |
| Verdict | Runs |
| Max context | 32K |
| Estimated speed | ~225 tok/s |
EXAONE 3.5 2.4B on the NVIDIA L40S: VRAM by quantization
| Quant | VRAM needed | Fits 48 GB? | Max context | Whole PC |
|---|---|---|---|---|
| F16 | 6.2 GB | ✓ Yes | 32K | — |
| Q8_0 | 4 GB | ✓ Yes | 32K | — |
| Q6_K | 3.4 GB | ✓ Yes | 32K | — |
| Q5_K_M | 3.1 GB | ✓ Yes | 32K | — |
| Q4_K_M | 2.9 GB | ✓ Yes | 32K | — |
| Q3_K_M | 2.5 GB | ✓ Yes | 32K | — |
| Q2_K | 2.2 GB | ✓ Yes | 32K | — |
Assumes an 8K-token context with an f16 KV cache. A longer window needs more; a quantized KV cache needs less. “Max context” is the largest window that still fits in 48 GB. Figures are estimates from parameter count, quantization and memory bandwidth — the analyzer lets you tune KV-cache quant and context.
Context costs VRAM too. At Q4_K_M on the NVIDIA L40S: 8K 2.9 GB · 32K 4.8 GB. Every window this model publishes still fits 48 GB.
What to do next
Nothing to change — run it at Q4_K_M. If you want more capability, check whether the next size up also fits.
No compatible GPU? EXAONE 3.5 2.4B on 8 GB of system RAM, CPU only: runs comfortably.
Running EXAONE 3.5 2.4B on this card:
At 2 hrs/day, buying (~$7,499) beats renting at $0.34/hr after about 30.6 years.
Affiliate links — we may earn a commission if you sign up, at no extra cost to you.
Cloud rates verified 2026-07 — estimates, and marketplace prices vary. Buying price is GPU MSRP only, not a full PC.
Other EXAONE 3.5 sizes on the NVIDIA L40S
- EXAONE 3.5 32B — runs (22.3 GB)
- EXAONE 3.5 7.8B — runs (6.6 GB)
EXAONE 3.5 2.4B on nearby hardware
- Apple M3 Pro (36 GB) — runs
All EXAONE 3.5 sizes on the NVIDIA L40S | EXAONE 3.5 2.4B model page | NVIDIA L40S GPU page | VRAM calculator | Check your hardware