Can the NVIDIA RTX 5090 run Mistral Small 24B?

RTX 5090 × Mistral Small 24B8K context
YesRuns fully on the GPU
Memory at Q8_027 GB of 32 GB
Weights 25.1 GBKV cache 1.3 GBRuntime 0.6 GB
QuantQ8_0Near-lossless
Memory27 GBof 32 GB usable
Speed~43tokens per second
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

Yes. The NVIDIA RTX 5090 can run Mistral Small 24B fully in VRAM at Q8_0 (near-lossless), using about 27 GB of the 32 GB available, at roughly 43 tokens/second.

Run it with Ollama

With Ollama installed, one command downloads and starts Mistral Small 24B:

ollama run mistral-small:24b

Ollama's default download is Q4_K_M (about 16.2 GB here). You have room for Q8_0, which is near-lossless: pick that GGUF in LM Studio or from Hugging Face.

Run it in the cloud

No NVIDIA RTX 5090? Run Mistral Small 24B in the cloud.

Rent a cloud GPU by the hour. No upfront cost, stop any time.

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 32 GB the NVIDIA RTX 5090 can use.

QuantQualityMemoryFits?Speed
FP16Full quality49.1 GBToo big—
Q8_0Near-lossless27 GBFits~43 tok/s
Q6_KExcellent21.3 GBFits~56 tok/s
Q5_K_MVery good18.7 GBFits~64 tok/s
Q4_K_MGood, most popular16.2 GBFits~75 tok/s
Q3_K_MNoticeably degraded13.5 GBFits~93 tok/s
Q2_KHeavily degraded10.8 GBFits~121 tok/s

Frequently asked questions

Can NVIDIA RTX 5090 run Mistral Small 24B?

Yes. The NVIDIA RTX 5090 can run Mistral Small 24B fully in VRAM at Q8_0 (near-lossless), using about 27 GB of the 32 GB available, at roughly 43 tokens/second.

How much VRAM does Mistral Small 24B need?

About 16.2 GB at Q4_K_M with 8,192 tokens of context, and about 49.1 GB at full FP16 precision.

How fast is Mistral Small 24B on the NVIDIA RTX 5090?

Roughly 43 tokens per second at Q8_0, estimated from the NVIDIA RTX 5090's 1792 GB/s memory bandwidth.