LL

llama3 42b v0 i1

LlamaGGUF by mradermacherUpdated 2024-05-05
Download on Hugging Face
Parameters43.2B
Context8K
Layers48
Architecturellama
Smallest quant9.6 GB
Quick answer

llama3 42b v0 i1 needs about 28.2 GB of memory at Q4_K_M (a 26.1 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 21 quants, from 9.6 GB (IQ1_S) to 35.4 GB (Q6_K).

Llama 3 42B pruned base model in GGUF format

Llama 3 42B v0 i1 is a pruned version of Meta’s Llama 3 70B. It contains 43.2 billion parameters and is available in multiple GGUF quantizations. The model is untrained and should be used with appropriate prompts.

Good at

  • Large context handling
  • High parameter count
  • LLM inference with GGUF

Limits

  • Untrained foundation model
  • No instruction tuning
  • No vision support

Best for

  • Research on pruning effects
  • Testing GGUF quantization
  • Benchmarking Llama 3 variants

Who it’s for. Researchers and developers exploring Llama 3 pruning and GGUF quantization. The model is suitable for those who need a large‑parameter base model for experimentation.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S9.6 GB 11.7 GB
Too big Get file →
IQ1_M10.5 GB 12.6 GB
Too big Get file →
IQ2_XXS11.9 GB 14.0 GB
Too big Get file →
IQ2_XS13.1 GB 15.2 GB
Too big Get file →
IQ2_S13.8 GB 15.9 GB
Too big Get file →
IQ2_M14.9 GB 17.1 GB
Too big Get file →
Q2_K16.3 GB 18.4 GB
Too big Get file →
IQ3_XXS16.9 GB 19.0 GB
Too big Get file →
IQ3_XS18.1 GB 20.2 GB
Too big Get file →
Q3_K_S19.0 GB 21.1 GB
Too big Get file →
IQ3_S19.0 GB 21.1 GB
Too big Get file →
IQ3_M19.6 GB 21.8 GB
Too big Get file →
Q3_K_M21.0 GB 23.2 GB
Too big Get file →
Q3_K_L22.8 GB 24.9 GB
Too big Get file →
IQ4_XS23.3 GB 25.4 GB
Too big Get file →
Q4_024.7 GB 26.8 GB
Too big Get file →
Q4_K_S24.7 GB 26.9 GB
Too big Get file →
Q4_K_M26.1 GB 28.2 GB
Too big Get file →
Q5_K_S29.8 GB 31.9 GB
Too big Get file →
Q5_K_M30.6 GB 32.7 GB
Too big Get file →
Q6_K35.4 GB 37.5 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (9.64 GB file, 11.7 GB memory). For 12 GB VRAM, IQ1_M (10.48 GB file, 12.6 GB memory). For 16 GB VRAM, IQ2_XS (13.12 GB file, 15.2 GB memory).

What GPU do you need for llama3 42b v0 i1?

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 IQ1_S (9.6 GB file, about 11.7 GB in use).
  • 16 GB VRAM: the best quant that fits is IQ2_S (13.8 GB file, about 15.9 GB in use).
  • 24 GB VRAM: the best quant that fits is Q3_K_M (21.0 GB file, about 23.2 GB in use).
  • 32 GB VRAM: the best quant that fits is Q5_K_S (29.8 GB file, about 31.9 GB in use).
  • 48 GB VRAM: the best quant that fits is Q6_K (35.4 GB file, about 37.5 GB in use).

Which llama3 42b v0 i1 quant should you choose?

Q4_K_M is the usual starting point for llama3 42b v0 i1: it balances file size and quality well. Here it is a 26.1 GB file.

If memory is tight, the smallest file is IQ1_S at 9.6 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
Parameters43.2B
Layers48
Trained context8,192 tokens
KV cache (fp16)192 KB per token

Source

Quantized bymradermacher
FamilyLlama
Licensellama3
Downloads (HF)999
Last updated2024-05-05

Frequently asked questions about llama3 42b v0 i1

How much VRAM does llama3 42b v0 i1 need?

About 28.2 GB at Q4_K_M with an 8K context window. That is the 26.1 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 llama3 42b v0 i1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 11.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of llama3 42b v0 i1 GGUF?

It depends on the quant: IQ1_S is 9.6 GB, IQ1_M is 10.5 GB, IQ2_XXS is 11.9 GB, IQ2_XS is 13.1 GB, IQ2_S is 13.8 GB, IQ2_M is 14.9 GB and 15 more.

What context length does llama3 42b v0 i1 support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 1.6 GB at 8K and 1.6 GB at 8K.

Where can I download llama3 42b v0 i1 GGUF?

The files are on Hugging Face in the mradermacher/llama3-42b-v0-i1-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 llama3 42b v0 i1 use?

Hugging Face lists the license as llama3. 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 8 GB VRAM?

Yes, the IQ1_S quantization requires 11.7 GB memory at 8 k context, which fits in 8 GB VRAM with some overhead.

Does this model support vision tasks?

No, the model has no vision support.

Is the model instruction tuned?

No, it is an untrained base model and should not be used with instruction prompts.

Other GGUFs of llama3-42b-v0

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Llama models

All Llama →

Other large models

All large models →