LL

Llama 3.1 Unhinged Vision 8B

LlamaGGUF by FiditeNeminiUpdated 2024-08-03
Download on Hugging Face
Parameters8.03B
Context8K
Layers32
Architecturellama
Smallest quant8.5 GB
Quick answer

Llama 3.1 Unhinged Vision 8B needs about 17.6 GB of memory at F16 (a 16.1 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 3 quants, from 8.5 GB (Q8_0) to 32.1 GB (F32).

8B Llama 3.1 Unhinged Vision model for multimodal tasks

Llama 3.1 Unhinged Vision 8B is a 8.03B parameter model with vision capability. It supports up to 8192 context tokens and is available in multiple quantizations. The model is a combination of a Meta Llama 3.1 8B base and a mmprojector for image understanding.

Good at

  • Vision and text multimodal dialogue
  • Handling two images in a single chat
  • Large context reasoning up to 8192 tokens

Limits

  • Requires 10.1GB VRAM for Q8_0 at 8k context
  • No vision support in lower quantizations
  • License is llama3.1

Best for

  • Chatting about multiple images
  • Image captioning in conversation
  • Multimodal content generation

Who it’s for. Developers building multimodal chatbots or content creation tools that need a lightweight 8B model. It is suitable for users with moderate VRAM who can accommodate the 10.1GB requirement for the Q8_0 quant.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_08.5 GB 10.1 GB
Too big Get file →
F1616.1 GB 17.6 GB
Too big Get file →
F3232.1 GB 33.7 GB
Too big Get file →

For 8GB VRAM, Q8_0 is too large; consider using Q8_0 on systems with at least 12GB VRAM. If you have 16GB VRAM, Q8_0 is suitable. For larger VRAM budgets, F16 or F32 provide higher precision but require 17.6GB and 33.7GB respectively.

What GPU do you need for Llama 3.1 Unhinged Vision 8B?

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: the best quant that fits is Q8_0 (8.5 GB file, about 10.1 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (8.5 GB file, about 10.1 GB in use).
  • 24 GB VRAM: the best quant that fits is F16 (16.1 GB file, about 17.6 GB in use).
  • 32 GB VRAM: the best quant that fits is F16 (16.1 GB file, about 17.6 GB in use).
  • 48 GB VRAM: the best quant that fits is F32 (32.1 GB file, about 33.7 GB in use).

Which Llama 3.1 Unhinged Vision 8B quant should you choose?

F16 (16.1 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.

If memory is tight, the smallest file is Q8_0 at 8.5 GB, with the quality trade-offs that come with fewer bits. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters8.03B
Layers32
Trained context8,192 tokens
KV cache (fp16)128 KB per token

Source

Quantized byFiditeNemini
FamilyLlama
Licensellama3.1
Downloads (HF)809
Last updated2024-08-03

Frequently asked questions about Llama 3.1 Unhinged Vision 8B

How much VRAM does Llama 3.1 Unhinged Vision 8B need?

About 17.6 GB at F16 with an 8K context window. That is the 16.1 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 Llama 3.1 Unhinged Vision 8B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 10.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Llama 3.1 Unhinged Vision 8B GGUF?

It depends on the quant: Q8_0 is 8.5 GB, F16 is 16.1 GB, F32 is 32.1 GB.

What context length does Llama 3.1 Unhinged Vision 8B support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 1.1 GB at 8K and 1.1 GB at 8K.

Where can I download Llama 3.1 Unhinged Vision 8B GGUF?

The files are on Hugging Face in the FiditeNemini/Llama-3.1-Unhinged-Vision-8B-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 Llama 3.1 Unhinged Vision 8B use?

Hugging Face lists the license as llama3.1. 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 this model on 8GB VRAM?

Q8_0 needs about 10.1GB VRAM at 8k context, so 8GB is insufficient. You would need more VRAM or a different quant.

Does it support vision?

Yes, the model has vision capability as indicated by detected abilities.

What is the maximum context length?

The model supports up to 8192 tokens of context.

Similar models

Browse all →

More Llama models

All Llama →

Other small models

All small models →