MI

Mistral Nemo Instruct 2407

MistralGGUF by second-stateUpdated 2024-07-24
Download on Hugging Face
Parameters12.2B
Context1000K
Layers40
Architecturellama
Smallest quant4.8 GB
Quick answer

Mistral Nemo Instruct 2407 needs about 9.3 GB of memory at Q4_K_M (a 7.5 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 13 quants, from 4.8 GB (Q2_K) to 24.5 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 5.4 GB.

Mistral Nemo Instruct 2407 – 12.2B parameter LLM with 128k context

Mistral Nemo Instruct 2407 is a 12.2B parameter transformer model trained with a 128k context window. It is an instruct fine‑tuned version of Mistral‑Nemo‑Base‑2407 and supports multilingual and code data. The model is released under Apache 2.0 and can be used with llama.cpp, transformers, or NeMo.

Good at

  • Large context handling
  • Multilingual instruction following
  • Code generation

Limits

  • No vision support
  • Requires 128k context window
  • No explicit GPU memory limits listed

Best for

  • Chatbots needing long context
  • Code assistance in multiple languages
  • Research on large‑context LLMs

Who it’s for. Developers and researchers who need a 12.2B parameter LLM with a 128k context window and multilingual capabilities. The model is suitable for applications that can accommodate the memory requirements of the chosen quantization.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K4.8 GB 6.6 GB
Fits Get file →
Q3_K_S5.5 GB 7.4 GB
Tight Get file →
Q3_K_M6.1 GB 7.9 GB
Tight Get file →
Q3_K_L6.6 GB 8.4 GB
Too big Get file →
Q4_07.1 GB 8.9 GB
Too big Get file →
Q4_K_S7.1 GB 9.0 GB
Too big Get file →
Q4_K_M7.5 GB 9.3 GB
Too big Get file →
Q5_08.5 GB 10.4 GB
Too big Get file →
Q5_K_S8.5 GB 10.4 GB
Too big Get file →
Q5_K_M8.7 GB 10.6 GB
Too big Get file →
Q6_K10.1 GB 11.9 GB
Too big Get file →
Q8_013.0 GB 14.9 GB
Too big Get file →
F1624.5 GB 26.3 GB
Too big Get file →

For 8 GB VRAM, the Q4_K quantization (file size 7.07 GB, estimated 8.9 GB memory at 8k context) is the closest fit. If more VRAM is available, Q5_K_S (8.52 GB, 10.4 GB memory) or Q6_K (10.06 GB, 11.9 GB memory) provide higher quality. Larger quants like Q8_0 or F16 require 14.9 GB or 26.3 GB memory respectively and are suitable for high‑end GPUs.

What GPU do you need for Mistral Nemo Instruct 2407?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: the best quant that fits is Q3_K_M (6.1 GB file, about 7.9 GB in use).
  • 12 GB VRAM: the best quant that fits is Q6_K (10.1 GB file, about 11.9 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (13.0 GB file, about 14.9 GB in use).
  • 24 GB VRAM: the best quant that fits is Q8_0 (13.0 GB file, about 14.9 GB in use).
  • 32 GB VRAM: the best quant that fits is F16 (24.5 GB file, about 26.3 GB in use).

Which Mistral Nemo Instruct 2407 quant should you choose?

Q4_K_M is the usual starting point for Mistral Nemo Instruct 2407: it balances file size and quality well. Here it is a 7.5 GB file.

The Q8_0 file is 13.0 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 4.8 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters12.2B
Layers40
Trained context1,024,000 tokens
KV cache (fp16)160 KB per token

Source

Quantized bysecond-state
FamilyMistral
Licenseapache-2.0
Downloads (HF)1,608
Last updated2024-07-24

Frequently asked questions about Mistral Nemo Instruct 2407

How much VRAM does Mistral Nemo Instruct 2407 need?

About 9.3 GB at Q4_K_M with an 8K context window. That is the 7.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Mistral Nemo Instruct 2407 on an 8 GB GPU?

Yes, with Q3_K_M: the 6.1 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Mistral Nemo Instruct 2407 GGUF?

It depends on the quant: Q2_K is 4.8 GB, Q3_K_S is 5.5 GB, Q3_K_M is 6.1 GB, Q3_K_L is 6.6 GB, Q4_0 is 7.1 GB, Q4_K_S is 7.1 GB and 7 more.

What context length does Mistral Nemo Instruct 2407 support?

The GGUF header lists a trained context length of 1,024,000 tokens (1000K). The KV cache grows with the context you use: about 1.3 GB at 8K and 5.4 GB at 32K.

Where can I download Mistral Nemo Instruct 2407 GGUF?

The files are on Hugging Face in the second-state/Mistral-Nemo-Instruct-2407-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Mistral Nemo Instruct 2407 use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Mistral Nemo Instruct 2407 on 8 GB VRAM?

The Q4_K quantization uses about 8.9 GB memory at 8k context, so it is borderline for 8 GB VRAM. A lower quant like Q2_K (6.6 GB) would fit more comfortably.

What context size does the model support?

The model was trained with a 128k context window and supports up to 128000 tokens.

Is the model licensed for commercial use?

It is released under the Apache 2.0 license, which permits commercial use.

Other GGUFs of Mistral-Nemo-Instruct-2407

Similar models

Browse all →

More from second-state

All by second-state →

More Mistral models

All Mistral →

Other medium models

All medium models →