FA

Falcon

FalconGGUF by mradermacherUpdated 2024-11-18
Download on Hugging Face
Parameters41.8B
Context2K
Layers60
Architecturefalcon
Smallest quant16.3 GB
Quick answer

Falcon needs about 27.5 GB of memory at Q4_K_M (a 26.0 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 14 quants, from 16.3 GB (Q2_K) to 45.0 GB (Q8_0).

Falcon-40B-Instruct: 41.8B parameter instruct model in GGUF format

Falcon-40B-Instruct is a 41.8B parameter causal decoder-only model fine‑tuned on a chat dataset. It is available under the Apache‑2.0 license and supports English and French. The model is optimized for inference with FlashAttention and multi‑query attention.

Good at

  • Chat and instruction following
  • Large context handling up to 2048 tokens
  • High parameter count for detailed responses

Limits

  • Requires large VRAM; at least 85‑100GB for full precision
  • No vision or multimodal support
  • Limited language coverage beyond English and French

Best for

  • Chatbot back‑ends needing high detail
  • Research on large‑scale language modeling
  • Enterprise instruction‑based applications

Who it’s for. Engineers and researchers who need a large, instruction‑tuned model for text generation and want to run it locally using GGUF. The model is suitable for those with sufficient GPU memory and who comply with the Apache‑2.0 license.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K16.3 GB 17.0 GB
Too big Get file →
IQ3_XS18.3 GB 19.0 GB
Too big Get file →
IQ3_S18.8 GB 19.6 GB
Too big Get file →
Q3_K_S18.8 GB 19.6 GB
Too big Get file →
IQ3_M19.7 GB 20.4 GB
Too big Get file →
Q3_K_M20.6 GB 21.3 GB
Too big Get file →
Q3_K_L22.1 GB 22.8 GB
Too big Get file →
IQ4_XS23.3 GB 24.1 GB
Too big Get file →
Q4_K_S24.3 GB 25.1 GB
Too big Get file →
Q4_K_M26.0 GB 26.7 GB
Too big Get file →
Q5_K_S29.5 GB 30.2 GB
Too big Get file →
Q5_K_M31.1 GB 31.9 GB
Too big Get file →
Q6_K35.0 GB 35.7 GB
Too big Get file →
Q8_045.0 GB 45.7 GB
Too big Get file →

For 8‑GB VRAM, use Q2_K (file size 16.26 GB, estimated 17.8 GB memory at 8k context). For 16‑GB VRAM, Q4_K_S (24.31 GB, 25.8 GB memory) or Q4_K_M (25.99 GB, 27.5 GB memory). For 24‑GB VRAM, Q5_K_S (29.47 GB, 31 GB memory) or Q5_K_M (31.14 GB, 32.6 GB memory).

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: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: the best quant that fits is Q3_K_L (22.1 GB file, about 23.6 GB in use).
  • 32 GB VRAM: the best quant that fits is Q5_K_S (29.5 GB file, about 31.0 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (45.0 GB file, about 46.5 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 26.0 GB file.

The Q8_0 file is 45.0 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 16.3 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
Parameters41.8B
Layers60
Trained context2,048 tokens
KV cache (fp16)120 KB per token

Source

Quantized bymradermacher
FamilyFalcon
Licenseapache-2.0
Downloads (HF)2,707
Last updated2024-11-18

Frequently asked questions about Falcon

How much VRAM does Falcon need?

About 27.5 GB at Q4_K_M with an 8K context window. That is the 26.0 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?

Not at an 8K context. The smallest file needs about 17.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Falcon GGUF?

It depends on the quant: Q2_K is 16.3 GB, IQ3_XS is 18.3 GB, IQ3_S is 18.8 GB, Q3_K_S is 18.8 GB, IQ3_M is 19.7 GB, Q3_K_M is 20.6 GB and 8 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 1.0 GB at 8K and 0.3 GB at 2K.

Where can I download Falcon GGUF?

The files are on Hugging Face in the mradermacher/falcon-40b-instruct-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 Falcon 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.

Can I run Falcon-40B-Instruct on 8 GB VRAM?

Use the Q2_K quant, which requires about 17.8 GB memory at 8k context, so 8 GB VRAM is insufficient for full inference.

What is the minimum memory needed for inference?

At least 85‑100 GB of memory is recommended for full‑precision inference; quantized versions reduce this requirement.

Does the model support vision tasks?

No, Falcon-40B-Instruct is a text‑only causal decoder‑only model.

Other GGUFs of falcon-40b-instruct

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Falcon models

All Falcon →

Other large models

All large models →