LL

Llama 3.1 8b Pruned 3 Layers

LlamaGGUF by mradermacherUpdated 2024-08-04
Download on Hugging Face
Parameters7.16B
Context128K
Layers28
Architecturellama
Smallest quant2.9 GB
Quick answer

Llama 3.1 8b Pruned 3 Layers needs about 5.8 GB of memory at Q4_K_M (a 4.4 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 15 quants, from 2.9 GB (Q2_K) to 14.3 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.

Llama 3.1 8b Pruned 3 Layers – lightweight 8‑billion parameter model

Llama 3.1 8b Pruned 3 Layers is a 7.16 billion parameter model with a 131072‑token context window. It is available in multiple GGUF quantizations ranging from Q2_K to F16. The model is unlicensed and has no specified language or vision capabilities.

Good at

  • Fast inference on modest GPUs
  • Low‑memory deployment
  • Short‑context tasks

Limits

  • No vision support
  • License not specified
  • Requires 8 k context for memory estimates

Best for

  • Chatbots on edge devices
  • Rapid prototyping
  • Low‑latency inference

Who it’s for. Developers and researchers needing a compact Llama 3.1 variant for quick experimentation or deployment on hardware with limited VRAM. The model is suitable for those who can work without a formal license and are comfortable with GGUF quantizations.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K2.9 GB 4.3 GB
Fits Get file →
IQ3_XS3.2 GB 4.6 GB
Fits Get file →
Q3_K_S3.3 GB 4.7 GB
Fits Get file →
IQ3_S3.3 GB 4.7 GB
Fits Get file →
IQ3_M3.4 GB 4.8 GB
Fits Get file →
Q3_K_M3.6 GB 5.0 GB
Fits Get file →
Q3_K_L3.9 GB 5.3 GB
Fits Get file →
IQ4_XS4.0 GB 5.4 GB
Fits Get file →
Q4_K_S4.2 GB 5.6 GB
Fits Get file →
Q4_K_M4.4 GB 5.8 GB
Fits Get file →
Q5_K_S5.0 GB 6.4 GB
Fits Get file →
Q5_K_M5.1 GB 6.6 GB
Fits Get file →
Q6_K5.9 GB 7.3 GB
Tight Get file →
Q8_07.6 GB 9.1 GB
Too big Get file →
F1614.3 GB 15.8 GB
Too big Get file →

For 8 GB VRAM, Q4_K_M (4.4 GB file, ~5.8 GB memory at 8 k context) is a good balance. If 12 GB VRAM is available, Q5_K_M (5.12 GB file, ~6.6 GB memory) offers higher quality. For the lowest memory footprint, Q2_K (2.86 GB file, ~4.3 GB memory) can be used on 6 GB VRAM systems.

What GPU do you need for Llama 3.1 8b Pruned 3 Layers?

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: the best quant that fits is Q4_K_M (4.4 GB file, about 5.8 GB in use).
  • 8 GB VRAM: the best quant that fits is Q6_K (5.9 GB file, about 7.3 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (7.6 GB file, about 9.1 GB in use).
  • 16 GB VRAM: the best quant that fits is F16 (14.3 GB file, about 15.8 GB in use).

Which Llama 3.1 8b Pruned 3 Layers quant should you choose?

Q4_K_M is the usual starting point for Llama 3.1 8b Pruned 3 Layers: it balances file size and quality well. Here it is a 4.4 GB file.

The Q8_0 file is 7.6 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 2.9 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
Parameters7.16B
Layers28
Trained context131,072 tokens
KV cache (fp16)112 KB per token

Source

Quantized bymradermacher
FamilyLlama
Downloads (HF)675
Last updated2024-08-04

Frequently asked questions about Llama 3.1 8b Pruned 3 Layers

How much VRAM does Llama 3.1 8b Pruned 3 Layers need?

About 5.8 GB at Q4_K_M with an 8K context window. That is the 4.4 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 8b Pruned 3 Layers on an 8 GB GPU?

Yes, with Q6_K: the 5.9 GB file needs about 7.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Llama 3.1 8b Pruned 3 Layers GGUF?

It depends on the quant: Q2_K is 2.9 GB, IQ3_XS is 3.2 GB, Q3_K_S is 3.3 GB, IQ3_S is 3.3 GB, IQ3_M is 3.4 GB, Q3_K_M is 3.6 GB and 9 more.

What context length does Llama 3.1 8b Pruned 3 Layers support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.

Where can I download Llama 3.1 8b Pruned 3 Layers GGUF?

The files are on Hugging Face in the mradermacher/Llama-3.1-8b-Pruned-3-Layers-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run this model on 8 GB VRAM?

Yes, the Q4_K_M quant uses about 5.8 GB memory at 8 k context, fitting within 8 GB VRAM.

Does this model support vision tasks?

No, the model has no vision capabilities.

What is the context length of this model?

The trained context length is 131072 tokens.

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Llama models

All Llama →

Other small models

All small models →