VOOZH about

URL: https://willitrunai.com/models/mistral-nemo-12b

โ‡ฑ Mistral Nemo 12B VRAM Requirements โ€” GPU Compatibility


๐Ÿ‘ Mistral
Mistral

Mistral Nemo 12B

Current
๐Ÿ‘ huggingface
HuggingFace๐Ÿ‘ ollama
Ollama
182.9KDownloads1.7KLikesJul 2024Released128K tokensContextApache 2.0License32 BasicQuality

Mistral Nemo 12B (12B parameters) requires approximately 11.6 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 14 GB of VRAM.

Get started

โ€” copy & paste to run locally

Copy-paste commands to run Mistral Nemo 12B on your machine.

Run

ollama run mistral-nemo

Quick specs

Parameters12B
Architecturedense
Context128K tokens
Modalitytext
Min RAM4.7 GB
Rec. RAM7.3 GB (Q4_K_M)
LicenseApache 2.0
FamilyMistral
โœ“ Chat

About this model

The Mistral-Nemo-Instruct-2407 Large Language Model (LLM) is an instruct fine-tuned version of the Mistral-Nemo-Base-2407. Trained jointly by Mistral AI and NVIDIA, it significantly outperforms existing models smaller or similar in size.

  • โ€ขReleased under the Apache 2 License
  • โ€ขPre-trained and instructed versions
  • โ€ขTrained with a 128k context window
  • โ€ขTrained on a large proportion of multilingual and code data
  • โ€ขDrop-in replacement of Mistral 7B

Related models

Your hardware

Detecting...

Quick picks

Best budgetB
RX 7600 XT 16GB~$329 โ€” 25 tok/s
๐Ÿ‘ NVIDIA
Best overallB
RTX 4080 Super 16GB~$999 โ€” 94 tok/s

Best hardware

Top picks for Mistral Nemo 12B

RTX 4080 Super 16GBB
16 GB
RTX 5080 16GBB
16 GB
RTX 5080 Laptop 16GBB
16 GB
RTX 5070 Ti 16GBB
16 GB
RTX 4070 Ti Super 16GBB
16 GB

Run this model

Mistral Nemo 12B on RTX 4080 Super 16GBMistral Nemo 12B on RTX 5080 16GBMistral Nemo 12B on RTX 5080 Laptop 16GB

Quantization options

VRAM estimates by quant level

No hardware detected โ€” fit column shows raw VRAM estimates

QuantBitsVRAMQualityFit
Q2_K
2
4.7 GB
Lowโ€”
Q3_K_S
3
5.9 GB
Lowโ€”
NVFP4
4
6.7 GB
Mediumโ€”
Q4_K_M
4
7.3 GB
Mediumโ€”
Q5_K_M
5
8.6 GB
Highโ€”
Q6_K
6
9.8 GB
Highโ€”
Q8_0
8
12.8 GB
Very Highโ€”
F16
16
24.6 GB
Maximumโ€”

Quality benchmarks

Mistral Nemo 12B benchmark scores

Benchmark verified

Reasoning

MMLU-Pro68.0%
GPQA Diamond5.4%
MATH-50012.7%
ARC Challengeโ€”

General

Chatbot Arenaโ€”
IFEval63.8%

Source: official ยท 2024-07-18

Hardware compatibility

Fit estimates across all hardware

Open calculator

Computing compatibility...

Memory breakdown

Reference: RTX 2060 6GB

Weights7.3 GB
KV Cache2.4 GB
Runtime1.2 GB
Headroom0.6 GB

Frequently asked questions

FAQ โ€” Mistral Nemo 12B

See also

Quantization GuideScoring MethodologyVRAM Calculator