SM

SmolVLM-256M-Instruct

SmolLMGGUF by MungertUpdated 2025-09-24
Download on Hugging Face
Parameters163M
Context8K
Layers30
Architecturellama
Smallest quant0.1 GB
Quick answer

SmolVLM-256M-Instruct needs about 0.8 GB of memory at Q4_K_M (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 21 quants, from 0.1 GB (Q4_0) to 0.3 GB (BF16).

SmolVLM‑256M‑Instruct – a compact vision‑enabled LLM

SmolVLM‑256M‑Instruct is a 0.163 billion‑parameter model based on SmolLM2. It supports vision tasks and instruction following. The model is available in multiple GGUF quantizations and is licensed under apache‑2.0.

Good at

  • Vision‑enabled instruction following
  • Compact on‑device inference
  • Low‑memory usage

Limits

  • No explicit vision‑model size limits
  • Requires 8 k context for estimates
  • No non‑English language support

Best for

  • Edge‑device image captioning
  • Small‑scale multimodal chat
  • Rapid prototyping of vision‑LLMs

Who it’s for. Developers building lightweight multimodal applications who need a small, vision‑capable LLM. The model is suitable for on‑device inference where memory is limited.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_00.1 GB 0.8 GB
Fits Get file →
IQ3_XXS0.1 GB 0.8 GB
Fits Get file →
IQ3_S0.1 GB 0.8 GB
Fits Get file →
IQ3_XS0.1 GB 0.8 GB
Fits Get file →
Q3_K_S0.1 GB 0.8 GB
Fits Get file →
IQ3_M0.1 GB 0.8 GB
Fits Get file →
Q4_10.1 GB 0.8 GB
Fits Get file →
Q3_K_M0.1 GB 0.8 GB
Fits Get file →
IQ4_XS0.1 GB 0.8 GB
Fits Get file →
IQ4_NL0.1 GB 0.8 GB
Fits Get file →
Q5_00.1 GB 0.8 GB
Fits Get file →
Q4_K_S0.1 GB 0.8 GB
Fits Get file →
Q4_K_M0.1 GB 0.8 GB
Fits Get file →
Q5_K_S0.1 GB 0.8 GB
Fits Get file →
Q5_10.1 GB 0.8 GB
Fits Get file →
Q5_K_M0.1 GB 0.8 GB
Fits Get file →
Q6_K_M0.2 GB 0.9 GB
Fits Get file →
Q8_00.2 GB 0.9 GB
Fits Get file →
BF160.3 GB 1.0 GB
Fits Get file →
F160.3 GB 1.0 GB
Fits Get file →
BF160.3 GB 1.0 GB
Fits Get file →

For 8 GB VRAM, use Q4_0 (0.09 GB) or IQ3_XXS (0.1 GB). For 4 GB VRAM, choose Q4_0 or IQ3_XXS. For 2 GB VRAM, Q4_0 is the smallest option. Higher precision models such as F16 or BF16 require more memory and are not recommended for low‑budget hardware.

What GPU do you need for SmolVLM-256M-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 BF16 (0.3 GB file, about 1.0 GB in use).

Which SmolVLM-256M-Instruct quant should you choose?

Q4_K_M is the usual starting point for SmolVLM-256M-Instruct: it balances file size and quality well. Here it is a 0.1 GB file.

The Q8_0 file is 0.2 GB, about 1.5 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 Q4_0 at 0.1 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters163M
Layers30
Trained context8,192 tokens
KV cache (fp16)23 KB per token

Source

Quantized byMungert
FamilySmolLM
Licenseapache-2.0
Downloads (HF)578
Last updated2025-09-24

Frequently asked questions about SmolVLM-256M-Instruct

How much VRAM does SmolVLM-256M-Instruct need?

About 0.8 GB at Q4_K_M with an 8K context window. That is the 0.1 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 SmolVLM-256M-Instruct on an 8 GB GPU?

Yes, with BF16: the 0.3 GB file needs about 1.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of SmolVLM-256M-Instruct GGUF?

It depends on the quant: Q4_0 is 0.1 GB, IQ3_XXS is 0.1 GB, IQ3_S is 0.1 GB, IQ3_XS is 0.1 GB, Q3_K_S is 0.1 GB, IQ3_M is 0.1 GB and 15 more.

What context length does SmolVLM-256M-Instruct support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.2 GB at 8K.

Where can I download SmolVLM-256M-Instruct GGUF?

The files are on Hugging Face in the Mungert/SmolVLM-256M-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 SmolVLM-256M-Instruct 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 SmolVLM‑256M‑Instruct on 4 GB VRAM?

Yes, using Q4_0 or IQ3_XXS quantizations, which estimate 0.8 GB memory at 8 k context.

Does the model support non‑English languages?

The model primarily understands and generates content in English; other languages are not guaranteed.

Is there a vision‑only mode?

The model includes vision capabilities but does not separate a dedicated vision‑only mode; it is designed for multimodal instruction following.

Other GGUFs of SmolLM2-135M-Instruct

Similar models

Browse all →

More from Mungert

All by Mungert →

More SmolLM models

All SmolLM →

Other tiny models

All tiny models →