TH

ThinkExtension 4x7

ThinkextensionGGUF by LoSboccaccUpdated 2024-01-14
Download on Hugging Face
Parameters24.2B
Context32K
Layers32
Architecturellama
Smallest quant8.8 GB
Quick answer

ThinkExtension 4x7 needs about 16.2 GB of memory at Q4_K_M (a 14.6 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 5 quants, from 8.8 GB (Q2_K) to 48.3 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

ThinkExtension 4x7 is a 24.2B-parameter Thinkextension model in GGUF format, quantized by LoSboccacc. Its trained context length is 32K tokens. The repository has 5 quant files, from 8.8 GB (Q2_K) to 48.3 GB.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K8.8 GB 10.4 GB
Too big Get file →
Q3_K_S10.4 GB 12.0 GB
Too big Get file →
Q3_K_M11.6 GB 13.2 GB
Too big Get file →
Q4_K_M14.6 GB 16.2 GB
Too big Get file →
—48.3 GB 49.9 GB
Too big Get file →

What GPU do you need for ThinkExtension 4x7?

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 Q3_K_S (10.4 GB file, about 12.0 GB in use).
  • 16 GB VRAM: the best quant that fits is Q3_K_M (11.6 GB file, about 13.2 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_K_M (14.6 GB file, about 16.2 GB in use).
  • 32 GB VRAM: the best quant that fits is Q4_K_M (14.6 GB file, about 16.2 GB in use).
  • 48 GB VRAM: the best quant that fits is Q4_K_M (14.6 GB file, about 16.2 GB in use).

Which ThinkExtension 4x7 quant should you choose?

Q4_K_M is the usual starting point for ThinkExtension 4x7: it balances file size and quality well. Here it is a 14.6 GB file.

If memory is tight, the smallest file is Q2_K at 8.8 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
Parameters24.2B (approximate, mixture of experts)
Layers32
Trained context32,768 tokens
KV cache (fp16)128 KB per token

Source

Quantized byLoSboccacc
Licenseapache-2.0
Downloads (HF)11
Last updated2024-01-14

Frequently asked questions about ThinkExtension 4x7

How much VRAM does ThinkExtension 4x7 need?

About 16.2 GB at Q4_K_M with an 8K context window. That is the 14.6 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 ThinkExtension 4x7 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 10.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of ThinkExtension 4x7 GGUF?

It depends on the quant: Q2_K is 8.8 GB, Q3_K_S is 10.4 GB, Q3_K_M is 11.6 GB, Q4_K_M is 14.6 GB, file is 48.3 GB.

What context length does ThinkExtension 4x7 support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download ThinkExtension 4x7 GGUF?

The files are on Hugging Face in the LoSboccacc/ThinkExtension-4x7 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does ThinkExtension 4x7 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.

Similar models

Browse all →

More from LoSboccacc

All by LoSboccacc →

Other medium models

All medium models →