Can I Run Codestral on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM)?
Superseded model. Codestral has been superseded by Devstral. This page is kept for reference; the newer family is a better starting point.
View Devstral →
Written by Jakub Rusinowski · Last updated May 29, 2024
Yes
Yes — Codestral 22B at Q3_K_M needs about 12.1 GB of the 16 GB usable on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM) (~3.9 GB spare), at ~59.9 tok/s (estimated), with room for about 16,384 tokens of context.
Confidence: high · Recommended quantization: Q3_K_M · Estimated speed: ~59.9 tok/s
RTX 5080 Desktop (16 GB VRAM, 32 GB RAM) — what it gives a model
| Usable memory for models | 16 GB |
| Memory bandwidth | 960 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
Codestral on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM): memory by quantization
| Quant | Memory needed | Fits 16 GB? | Max context | Est. speed | Download |
|---|
| F16 | 47.1 GB | ✗ No | — | — | 44.4 GB |
| Q8_0 | 26.3 GB | ✗ No | — | — | 23.6 GB |
| Q6_K | 20.9 GB | ✗ No | — | — | 18.2 GB |
| Q5_K_M | 18.4 GB | ✗ No | — | — | 15.7 GB |
| Q4_K_M | 16.1 GB | ✗ No | — | — | 13.4 GB |
| Q3_K_M | 12.1 GB | ✓ Yes | 16K | ~59.9 tok/s | 9.5 GB |
| Q2_K | 10 GB | ✓ Yes | 32K | ~73.1 tok/s | 7.3 GB |
What to watch out for
- Q3_K_M is the only quantization that fits, and it is a heavily degraded one — expect noticeably worse output than the same model at Q4_K_M. A smaller model at Q4 is usually the better trade.
- 1 larger variant of Codestral does not fit and would need CPU offload or different hardware.
RTX 5080 desktop limitations
- 16 GB VRAM is the binding constraint, not compute — a slower 24 GB card runs strictly more models.
Recommended setup
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
How these numbers are calculated
- Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
- 16 GB of VRAM on the NVIDIA GeForce RTX 5080 at 960 GB/s.
- 32 GB of system RAM available for CPU offload when a model exceeds VRAM.
- Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
- KV cache is computed from this model's published attention configuration.
FAQ
Can I run Codestral on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM)?
Yes — Codestral 22B at Q3_K_M needs about 12.1 GB of the 16 GB usable on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM) (~3.9 GB spare), at ~59.9 tok/s (estimated), with room for about 16,384 tokens of context.
Which quantization of Codestral should I use on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM)?
Q3_K_M — it needs about 12.1 GB of the 16 GB available, downloads as roughly 9.5 GB, and runs at an estimated 59.9 tokens/sec with up to 16K of context.
What limits Codestral on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM)?
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Which runtime should I use?
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Other Computers
Other Models on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM)
Codestral on GPUs
What This Model Is Good At
Model & Tools
← Can I Run It? | Codestral model page | Check your hardware