Can I Run Cogito v1 on NVIDIA L40S?
Autor: Jakub Rusinowski · Ostatnia aktualizacja: 21 lipca 2026
Yes, but it's tight — Cogito v1 70B at Q4_K_M needs 45.7 GB at 8K context (42.3 GB weights + 3.4 GB KV cache/overhead), leaving ~2.3 GB of the NVIDIA L40S's 48 GB, at ~18 tok/s (est.), with room for up to 8K context.
See what else this hardware can run →
Jako partner Amazon zarabiamy na kwalifikujących się zakupach. Linki do chmurowych GPU są linkami polecającymi — możemy otrzymać prowizję bez dodatkowych kosztów dla Ciebie.
NVIDIA L40S Specs
| VRAM | 48 GB |
| Memory Bandwidth | 864 GB/s |
Cogito v1 70B on the NVIDIA L40S: VRAM by quantization
| Quant | VRAM needed | Fits 48 GB? | Max context | Whole PC |
|---|---|---|---|---|
| F16 | 143.5 GB | ✗ No | — | — |
| Q8_0 | 77.9 GB | ✗ No | — | — |
| Q6_K | 60.9 GB | ✗ No | — | — |
| Q5_K_M | 53.1 GB | ✗ No | — | — |
| Q4_K_M | 45.7 GB | ✓ Yes | 8K | — |
| Q3_K_M | 33.3 GB | ✓ Yes | 32K | — |
| Q2_K | 26.5 GB | ✓ Yes | 64K | — |
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 45.7 GB · 32K 53.8 GB ✗ · 128K 86 GB ✗. Past 32K it no longer fits 48 GB — a q8_0 KV cache buys roughly half of that back, which the calculator will price for you.
No compatible GPU? Cogito v1 on 8 GB of system RAM, CPU only: runs.
Cogito v1 Sizes That Fit the NVIDIA L40S (at 8K context)
| Cogito v1 70B | Q4_K_M · 45.7 GB at 8K context · ~18 tok/s (est.) |
| Cogito v1 32B | Q4_K_M · 22.3 GB at 8K context · ~36 tok/s (est.) |
| Cogito v1 14B | Q4_K_M · 10.9 GB at 8K context · ~72 tok/s (est.) |
| Cogito v1 8B | Q4_K_M · 6.7 GB at 8K context · ~112 tok/s (est.) |
| Cogito v1 3B | Q4_K_M · 3.6 GB at 8K context · ~196 tok/s (est.) |
At 2 hrs/day, buying (~$7,499) beats renting at $0.77/hr after about 13.5 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.
FAQ
Does the NVIDIA L40S have enough VRAM for Cogito v1?
Yes, but it's tight — Cogito v1 70B at Q4_K_M needs 45.7 GB at 8K context (42.3 GB weights + 3.4 GB KV cache/overhead), leaving ~2.3 GB of the NVIDIA L40S's 48 GB, at ~18 tok/s (est.), with room for up to 8K context.
Which quantization of Cogito v1 should I use on the NVIDIA L40S?
Cogito v1 70B at Q4_K_M quantization needs 45.7 GB at 8K context (42.3 GB weights + 3.4 GB KV cache/overhead), estimated ~18 tokens/sec, up to 8K context.
Every Cogito v1 size on the NVIDIA L40S
| Size | VRAM at 8K | Verdict | Speed |
|---|---|---|---|
| Cogito v1 70B | 45.7 GB | Runs | ~18 tok/s |
| Cogito v1 32B | 22.3 GB | Runs | ~36 tok/s |
| Cogito v1 14B | 10.9 GB | Runs | ~72 tok/s |
| Cogito v1 8B | 6.7 GB | Runs | ~112 tok/s |
| Cogito v1 3B | 3.6 GB | Runs | ~196 tok/s |
Cogito v1 on Other GPUs
- Cogito v1 on NVIDIA RTX 6000 Ada Generation
- Cogito v1 on Apple M3 Pro
- Cogito v1 on NVIDIA GeForce RTX 5090
- Cogito v1 on Apple M4
- Cogito v1 on Apple M2 Pro
Popular Models on the NVIDIA L40S
VRAM Tier
Troubleshooting
- CUDA out of memory — why it happens and how to fix it
- Which GGUF quant should I download? (Q4 vs Q5 vs Q8)
Buying Guide
← Can I Run It? | Cogito v1 Model Page | NVIDIA L40S GPU Page | VRAM calculator | Check Your Hardware