Skip to content
Run LLM Locally

Mistral Nemo 12B Instruct on the RTX 3090 Ti

With 24 GB of VRAM, the RTX 3090 Ti runs Mistral Nemo 12B Instruct at Q4_K_M entirely on the GPU with an 8K context (about 9.4 GB needed). At Q8_0 it fits entirely.

ShareFacebookXWhatsAppLinkedInEmail

Mistral Nemo 12B Instruct

Q4_K_M · 8K context

Excellent

The model needs about 9.4 GB and fits in the 23.3 GB available, with 13.8 GB to spare.

Your hardware
RTX 3090 Ti 24 GB · 64 GB RAM
Estimated memory
~9.4 GB estimate
VRAM (usable memory)
23.3 GB
GPU only
Yes
CPU/GPU offload
Not required
Generation speed
78–111 tok/s estimate
Largest context in fast memory
92K tokens
Confidence: low

Based on: memory bandwidth (1,008 GB/s) and model size

Memory breakdown

~9.4 GB (9 GB–10 GB) needed · estimate

Fits entirely in VRAM · 13.8 GB left

  • Model weights7 GB
  • KV cache1.3 GB
  • Runtime overhead0.8 GB
  • Safety margin0.4 GB
Assumptions: context 8K tokens · KV cache f16 · llama.cpp (CUDA) · flash attention on · weights = real file size

EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card

What makes this estimate less reliable

  • no benchmark for this hardware and model yet

Other quantizations on the RTX 3090 Ti

Mistral Nemo 12B Instruct on RTX 3090 Ti + 64 GB RAM · 8K context
Mistral Nemo 12B Instruct Q8_08K~14.9 GB · 50–69 tok/s estimateExcellent
Mistral Nemo 12B Instruct Q6_K8K~12 GB · 62–86 tok/s estimateExcellent
Mistral Nemo 12B Instruct Q5_K_M8K~10.7 GB · 69–98 tok/s estimateExcellent
Mistral Nemo 12B Instruct Q4_K_M8K~9.4 GB · 78–111 tok/s estimateExcellent
Mistral Nemo 12B Instruct IQ4_XS8K~8.7 GB · 84–121 tok/s estimateExcellent
Mistral Nemo 12B Instruct Q3_K_M8K~8.1 GB · 90–132 tok/s estimateExcellent

Alternatives

GPUs that run Mistral Nemo 12B Instruct Q4_K_M entirely in VRAM

Other models that fit the RTX 3090 Ti

Where to buy a RTX 3090 Ti

Prices and links follow the market you choose.

GeForce RTX 3090 Ti

24 GB GDDR6X · 1,008 GB/s · 450 W

ExcellentRuns Mistral Nemo 12B Instruct Q4_K_M entirely in memory

Price not available

Availability unknown

View on amazon.com

Affiliate link

No current offer imported for this market: the link opens a marketplace search.

Frequently asked questions

Can the RTX 3090 Ti run Mistral Nemo 12B Instruct?
The model needs about 9.4 GB and fits in the 23.3 GB available, with 13.8 GB to spare.
Which quantization should I use on the RTX 3090 Ti?
Q8_0 is the highest-precision quantization that fits entirely in VRAM at 8K context. ~14.9 GB / 23.3 GB.
What is the largest context on the RTX 3090 Ti?
At Q4_K_M, about 92K tokens fit entirely in VRAM with an f16 KV cache. A q8_0 KV cache roughly doubles that.
How fast is Mistral Nemo 12B Instruct (RTX 3090 Ti)?
We estimate 78–111 tok/s (estimate, confidence low). Based on: memory bandwidth (1,008 GB/s) and model size.