MI

Fireball Mistral Nemo Base 2407 V2 f16

MistralGGUF by EpistemeAIUpdated 2024-08-20
Download on Hugging Face
Parameters12.2B
Context1000K
Layers40
Architecturellama
Smallest quant24.5 GB
Quick answer

Fireball Mistral Nemo Base 2407 V2 f16 needs about 26.3 GB of memory at F16 (a 24.5 GB file) with an 8K context window, so it fits on a 32 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 5.4 GB.

Fireball Mistral Nemo Base 2407 V2 f16 is a 12.2B-parameter Mistral model in GGUF format, quantized by EpistemeAI. Its trained context length is 1000K tokens. The repository has 1 quant file, from 24.5 GB (F16) to 24.5 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
F1624.5 GB 26.3 GB
Too big Get file →

What GPU do you need for Fireball Mistral Nemo Base 2407 V2 f16?

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: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: the best quant that fits is F16 (24.5 GB file, about 26.3 GB in use).

Which Fireball Mistral Nemo Base 2407 V2 f16 quant should you choose?

F16 (24.5 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

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 byEpistemeAI
FamilyMistral
Licenseapache-2.0
Downloads (HF)16
Last updated2024-08-20

Frequently asked questions about Fireball Mistral Nemo Base 2407 V2 f16

How much VRAM does Fireball Mistral Nemo Base 2407 V2 f16 need?

About 26.3 GB at F16 with an 8K context window. That is the 24.5 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 Fireball Mistral Nemo Base 2407 V2 f16 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 26.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Fireball Mistral Nemo Base 2407 V2 f16 GGUF?

It depends on the quant: F16 is 24.5 GB.

What context length does Fireball Mistral Nemo Base 2407 V2 f16 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 Fireball Mistral Nemo Base 2407 V2 f16 GGUF?

The files are on Hugging Face in the EpistemeAI/Fireball-Mistral-Nemo-Base-2407-V2-f16-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 Fireball Mistral Nemo Base 2407 V2 f16 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.

Other GGUFs of Fireball-Mistral-Nemo-Base-2407-sft-v2.2a

Similar models

Browse all →

More from EpistemeAI

All by EpistemeAI →

More Mistral models

All Mistral →

Other medium models

All medium models →