SmolLM2 135M Instruct heretic 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 15 quants, from 0.1 GB (IQ3_S) to 0.3 GB (F16).
SmolLM2 135M Instruct heretic – a lightweight, decensored LLM for edge deployment
SmolLM2 135M Instruct heretic is a 0.135B parameter transformer decoder model. It is a decensored variant of SmolLM2‑135M‑Instruct with refusal guardrails removed. The model runs on CPU‑only or low‑VRAM GPUs and is suitable for on‑device experiments.
Good at
- Instruction following
- Text rewriting
- Summarization
Limits
- English‑only content
- No vision support
- Requires ~0.8 GB VRAM at 8 k context for most quants
Best for
- Edge inference
- Browser demos
- Raspberry Pi projects
Who it’s for. This model is intended for developers and researchers needing a small, decensored LLM for CPU‑only inference or low‑VRAM GPUs. It is useful for studying refusal mechanisms or deploying on embedded devices.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ3_S | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q2_K | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q3_K_S | 0.1 GB | 0.8 GB | Fits | Get file → | |
| IQ4_XS | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q4_0 | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q3_K_M | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q3_K_L | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q4_1 | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q4_K_S | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q4_K_M | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q5_K_S | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q5_K_M | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q6_K | 0.1 GB | 0.8 GB | Fits | Get file → | |
| Q8_0 | 0.1 GB | 0.8 GB | Fits | Get file → | |
| F16 | 0.3 GB | 1.0 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K_M or Q4_K_S (0.1 GB file, ~0.8 GB memory). For 4 GB VRAM, drop to Q3_K_S or IQ3_S (0.09 GB file, ~0.8 GB memory). If memory is very tight, use IQ4_XS (0.09 GB file).
What GPU do you need for SmolLM2 135M Instruct heretic?
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 F16 (0.3 GB file, about 1.0 GB in use).
Which SmolLM2 135M Instruct heretic quant should you choose?
Q4_K_M is the usual starting point for SmolLM2 135M Instruct heretic: it balances file size and quality well. Here it is a 0.1 GB file.
The Q8_0 file is 0.1 GB, about 1.4 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 IQ3_S at 0.1 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about SmolLM2 135M Instruct heretic
How much VRAM does SmolLM2 135M Instruct heretic 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 SmolLM2 135M Instruct heretic on an 8 GB GPU?
Yes, with F16: 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 SmolLM2 135M Instruct heretic GGUF?
It depends on the quant: IQ3_S is 0.1 GB, Q2_K is 0.1 GB, Q3_K_S is 0.1 GB, IQ4_XS is 0.1 GB, Q4_0 is 0.1 GB, Q3_K_M is 0.1 GB and 9 more.
What context length does SmolLM2 135M Instruct heretic 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 SmolLM2 135M Instruct heretic GGUF?
The files are on Hugging Face in the saidutta69/SmolLM2-135M-Instruct-heretic repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does SmolLM2 135M Instruct heretic 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 SmolLM2 135M Instruct heretic on 8 GB VRAM?
Yes, most quants (e.g., Q4_K_M) use about 0.8 GB VRAM at 8 k context, leaving headroom on an 8 GB GPU.
Does the model support image input?
No, the model is a text‑only transformer and has no vision capability.
Is the model licensed for commercial use?
It is released under the Apache‑2.0 license, which permits commercial use.