FA

Falcon H1 7B Instruct

FalconGGUF by tiiuaeUpdated 2025-07-31
Download on Hugging Face
Parameters7.59B
Context256K
Layers44
Architecturefalcon-h1
Smallest quant1.9 GB
Quick answer

Falcon H1 7B Instruct needs about 5.5 GB of memory at Q4_K_M (a 4.6 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 34 quants, from 1.9 GB (IQ1_M) to 30.3 GB (F32). Longer context needs more: at 32K tokens the KV cache alone adds about 1.5 GB.

Falcon H1 7B Instruct – a 7.59B parameter causal decoder model

Falcon H1 7B Instruct is a causal decoder-only model built on a hybrid Transformers + Mamba architecture. It supports English and multilingual text generation. The model is released under the Falcon-LLM License.

Good at

  • Large context handling
  • Multilingual text generation
  • Reasoning tasks

Limits

  • No vision support
  • Requires 262144 token context
  • License not public domain

Best for

  • Chatbots needing long context
  • Language translation services
  • Content generation for multilingual sites

Who it’s for. Developers building conversational agents or translation tools that need a large context window. The model is suitable for users who can manage the 262144 token context and have the necessary GPU memory for inference.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_M1.9 GB 2.8 GB
Fits Get file →
TQ1_02.0 GB 2.9 GB
Fits Get file →
IQ2_XXS2.2 GB 3.1 GB
Fits Get file →
TQ2_02.3 GB 3.2 GB
Fits Get file →
IQ2_XS2.4 GB 3.3 GB
Fits Get file →
IQ2_S2.5 GB 3.4 GB
Fits Get file →
IQ2_M2.7 GB 3.6 GB
Fits Get file →
Q2_K_S2.7 GB 3.6 GB
Fits Get file →
Q2_K2.9 GB 3.8 GB
Fits Get file →
IQ3_XXS3.1 GB 3.9 GB
Fits Get file →
IQ3_XS3.3 GB 4.2 GB
Fits Get file →
Q3_K_S3.4 GB 4.3 GB
Fits Get file →
IQ3_S3.4 GB 4.3 GB
Fits Get file →
IQ3_M3.5 GB 4.4 GB
Fits Get file →
Q3_K3.7 GB 4.6 GB
Fits Get file →
Q3_K_M3.7 GB 4.6 GB
Fits Get file →
Q3_K_L3.9 GB 4.8 GB
Fits Get file →
IQ4_XS4.2 GB 5.0 GB
Fits Get file →
Q4_04.4 GB 5.2 GB
Fits Get file →
IQ4_NL4.4 GB 5.3 GB
Fits Get file →
Q4_K_S4.4 GB 5.3 GB
Fits Get file →
Q4_K4.6 GB 5.5 GB
Fits Get file →
Q4_K_M4.6 GB 5.5 GB
Fits Get file →
Q4_14.8 GB 5.7 GB
Fits Get file →
Q5_05.3 GB 6.1 GB
Fits Get file →
Q5_K_S5.3 GB 6.1 GB
Fits Get file →
Q5_K5.4 GB 6.3 GB
Fits Get file →
Q5_K_M5.4 GB 6.3 GB
Fits Get file →
Q5_15.7 GB 6.6 GB
Fits Get file →
Q6_K6.2 GB 7.1 GB
Fits Get file →
Q8_08.1 GB 8.9 GB
Too big Get file →
BF1615.2 GB 16.0 GB
Too big Get file →
F1615.2 GB 16.0 GB
Too big Get file →
F3230.3 GB 31.2 GB
Too big Get file →

For 8 GB VRAM, use IQ1_M (1.94 GB) or TQ1_0 (1.99 GB). For 12 GB VRAM, IQ2_M (2.68 GB) or Q2_K (2.89 GB) are good choices. For 16 GB VRAM, Q4_K (4.6 GB) or Q5_K (5.39 GB) provide higher quality while staying within memory limits. Estimates are approximate.

What GPU do you need for Falcon H1 7B Instruct?

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: the best quant that fits is IQ3_XXS (3.1 GB file, about 3.9 GB in use).
  • 6 GB VRAM: the best quant that fits is Q4_1 (4.8 GB file, about 5.7 GB in use).
  • 8 GB VRAM: the best quant that fits is Q6_K (6.2 GB file, about 7.1 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (8.1 GB file, about 8.9 GB in use).
  • 16 GB VRAM: the best quant that fits is BF16 (15.2 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is BF16 (15.2 GB file, about 16.0 GB in use).
  • 32 GB VRAM: the best quant that fits is F32 (30.3 GB file, about 31.2 GB in use).

Which Falcon H1 7B Instruct quant should you choose?

Q4_K_M is the usual starting point for Falcon H1 7B Instruct: it balances file size and quality well. Here it is a 4.6 GB file.

The Q8_0 file is 8.1 GB, about 1.8 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 IQ1_M at 1.9 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-h1
Parameters7.59B
Layers44
Trained context262,144 tokens
KV cache (fp16)44 KB per token

Source

Quantized bytiiuae
FamilyFalcon
Licenseother
Downloads (HF)2,893
Last updated2025-07-31

Frequently asked questions about Falcon H1 7B Instruct

How much VRAM does Falcon H1 7B Instruct need?

About 5.5 GB at Q4_K_M with an 8K context window. That is the 4.6 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 Falcon H1 7B Instruct on an 8 GB GPU?

Yes, with Q6_K: the 6.2 GB file needs about 7.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Falcon H1 7B Instruct GGUF?

It depends on the quant: IQ1_M is 1.9 GB, TQ1_0 is 2.0 GB, IQ2_XXS is 2.2 GB, TQ2_0 is 2.3 GB, IQ2_XS is 2.4 GB, IQ2_S is 2.5 GB and 28 more.

What context length does Falcon H1 7B Instruct support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.5 GB at 32K.

Where can I download Falcon H1 7B Instruct GGUF?

The files are on Hugging Face in the tiiuae/Falcon-H1-7B-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 H1 7B Instruct use?

Hugging Face lists the license as other. 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 H1 7B Instruct on 8 GB VRAM?

Yes, use the IQ1_M or TQ1_0 quantized files, which are under 2 GB and require about 3 GB of VRAM at 8 k context.

Does the model support image inputs?

No, Falcon H1 7B Instruct is a text-only model and does not support vision tasks.

What context length does the model support?

The model was trained with a context of 262144 tokens and can handle that length during inference.

Other GGUFs of Falcon-H1-7B-Instruct

Similar models

Browse all →

More from tiiuae

All by tiiuae →

More Falcon models

All Falcon →

Other small models

All small models →