FA

Falcon

FalconGGUF by CharlieJiUpdated 2024-01-29
Download on Hugging Face
Parameters7.22B
Context2K
Layers32
Architecturefalcon
Smallest quant3.9 GB
Quick answer

Falcon needs about 5.5 GB of memory at Q4_K_M (a 5.0 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 9 quants, from 3.9 GB (Q2_K) to 7.0 GB (Q6_K).

Falcon is a 7.22B-parameter Falcon model in GGUF format, quantized by CharlieJi. Its trained context length is 2K tokens. The repository has 9 quant files, from 3.9 GB (Q2_K) to 7.0 GB (Q6_K).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K3.9 GB 4.4 GB
Fits Get file →
Q3_K_S4.1 GB 4.6 GB
Fits Get file →
Q3_K_M4.4 GB 4.9 GB
Fits Get file →
Q3_K_L4.6 GB 5.1 GB
Fits Get file →
Q4_K_S4.7 GB 5.3 GB
Fits Get file →
Q4_K_M5.0 GB 5.5 GB
Fits Get file →
Q5_K_S5.3 GB 5.9 GB
Fits Get file →
Q5_K_M5.7 GB 6.2 GB
Fits Get file →
Q6_K7.0 GB 7.5 GB
Tight Get file →

What GPU do you need for Falcon?

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 Q5_K_S (5.3 GB file, about 5.9 GB in use).
  • 8 GB VRAM: the best quant that fits is Q6_K (7.0 GB file, about 7.6 GB in use).

Which Falcon quant should you choose?

Q4_K_M is the usual starting point for Falcon: it balances file size and quality well. Here it is a 5.0 GB file.

If memory is tight, the smallest file is Q2_K at 3.9 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

Architecturefalcon
Parameters7.22B
Layers32
Trained context2,048 tokens
KV cache (fp16)8 KB per token

Source

Quantized byCharlieJi
FamilyFalcon
Downloads (HF)113
Last updated2024-01-29

Frequently asked questions about Falcon

How much VRAM does Falcon need?

About 5.5 GB at Q4_K_M with an 8K context window. That is the 5.0 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 Falcon on an 8 GB GPU?

Yes, with Q6_K: the 7.0 GB file needs about 7.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Falcon GGUF?

It depends on the quant: Q2_K is 3.9 GB, Q3_K_S is 4.1 GB, Q3_K_M is 4.4 GB, Q3_K_L is 4.6 GB, Q4_K_S is 4.7 GB, Q4_K_M is 5.0 GB and 3 more.

What context length does Falcon support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.0 GB at 2K.

Where can I download Falcon GGUF?

The files are on Hugging Face in the CharlieJi/gorilla-falcon-7b-hf-v0-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Similar models

Browse all →

More from CharlieJi

All by CharlieJi →

More Falcon models

All Falcon →

Other small models

All small models →