QE

QED-Nano

QedGGUF by MungertUpdated 2026-02-23
Download on Hugging Face
Parameters4.02B
Context256K
Layers36
Architectureqwen3
Smallest quant0.0 GB
Quick answer

QED-Nano needs about 4.2 GB of memory at Q4_K_M (a 2.5 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 31 quants, from 0.0 GB to 8.1 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.8 GB.

4B parameter Qwen3 model fine‑tuned for mathematical reasoning

QED‑Nano is a 4.02B parameter model based on Qwen3. It is fine‑tuned for theorem proving and explicit chain‑of‑thought reasoning. The model is released under Apache‑2.0 and supports reasoning and tools.

Good at

  • Theorem proving
  • Explicit chain‑of‑thought reasoning
  • Tool usage

Limits

  • No vision support
  • Requires large VRAM for higher precision
  • License is Apache‑2.0

Best for

  • Math competition problem solving
  • Proof generation
  • Educational tutoring in mathematics

Who it’s for. Researchers and developers needing a compact, open‑source model for mathematical reasoning tasks. The model is suitable for integration into agents that require explicit reasoning steps.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 1.7 GB
Fits Get file →
IQ3_XXS1.8 GB 3.5 GB
Fits Get file →
IQ3_XS1.9 GB 3.6 GB
Fits Get file →
IQ3_M2.0 GB 3.7 GB
Fits Get file →
IQ3_S2.0 GB 3.7 GB
Fits Get file →
Q3_K_S2.2 GB 3.9 GB
Fits Get file →
Q3_K_M2.2 GB 3.9 GB
Fits Get file →
Q4_02.3 GB 4.0 GB
Fits Get file →
IQ4_XS2.3 GB 4.0 GB
Fits Get file →
Q3_K_L2.3 GB 4.0 GB
Fits Get file →
IQ4_NL2.4 GB 4.1 GB
Fits Get file →
Q4_K_S2.5 GB 4.2 GB
Fits Get file →
Q4_02.5 GB 4.2 GB
Fits Get file →
Q4_12.5 GB 4.2 GB
Fits Get file →
Q4_K_M2.5 GB 4.2 GB
Fits Get file →
Q4_K_L2.6 GB 4.3 GB
Fits Get file →
Q4_12.7 GB 4.4 GB
Fits Get file →
Q5_02.8 GB 4.5 GB
Fits Get file →
Q5_K_S2.9 GB 4.6 GB
Fits Get file →
Q5_02.9 GB 4.6 GB
Fits Get file →
Q5_K_M3.0 GB 4.7 GB
Fits Get file →
Q5_13.0 GB 4.7 GB
Fits Get file →
Q5_K_L3.0 GB 4.8 GB
Fits Get file →
Q5_13.1 GB 4.9 GB
Fits Get file →
Q6_K_M3.3 GB 5.0 GB
Fits Get file →
Q6_K_L3.4 GB 5.1 GB
Fits Get file →
Q8_04.3 GB 6.0 GB
Fits Get file →
BF165.9 GB 7.6 GB
Tight Get file →
F165.9 GB 7.6 GB
Tight Get file →
BF168.1 GB 9.8 GB
Too big Get file →
F168.1 GB 9.8 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_M (2.53 GB file, 4.2 GB memory at 8k context). If 12 GB VRAM is available, Q5_K_M (2.95 GB file, 4.7 GB memory) offers higher quality. For the highest fidelity, Q8_0 (4.28 GB file, 6 GB memory) can be used on 16 GB VRAM systems.

What GPU do you need for QED-Nano?

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 Q3_K_L (2.3 GB file, about 4.0 GB in use).
  • 6 GB VRAM: the best quant that fits is Q8_0 (4.3 GB file, about 6.0 GB in use).
  • 8 GB VRAM: the best quant that fits is BF16 (5.9 GB file, about 7.6 GB in use).
  • 12 GB VRAM: the best quant that fits is BF16 (8.1 GB file, about 9.8 GB in use).

Which QED-Nano quant should you choose?

Q4_K_M is the usual starting point for QED-Nano: it balances file size and quality well. Here it is a 2.5 GB file.

The Q8_0 file is 4.3 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 the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen3
Parameters4.02B
Layers36
Trained context262,144 tokens
KV cache (fp16)144 KB per token

Source

Quantized byMungert
FamilyQed
Licenseapache-2.0
Downloads (HF)538
Last updated2026-02-23

Frequently asked questions about QED-Nano

How much VRAM does QED-Nano need?

About 4.2 GB at Q4_K_M with an 8K context window. That is the 2.5 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 QED-Nano on an 8 GB GPU?

Yes, with BF16: the 5.9 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 QED-Nano GGUF?

It depends on the quant: file is 0.0 GB, IQ3_XXS is 1.8 GB, IQ3_XS is 1.9 GB, IQ3_M is 2.0 GB, IQ3_S is 2.0 GB, Q3_K_S is 2.2 GB and 25 more.

What context length does QED-Nano support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.2 GB at 8K and 4.8 GB at 32K.

Where can I download QED-Nano GGUF?

The files are on Hugging Face in the Mungert/QED-Nano-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 QED-Nano 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 QED‑Nano on 8 GB VRAM?

Yes, the Q4_K_M quantization requires about 4.2 GB memory at 8k context, which fits within 8 GB VRAM.

Does QED‑Nano support vision tasks?

No, the model does not include vision capabilities.

Similar models

Browse all →

More from Mungert

All by Mungert →

More Qed models

All Qed →

Other small models

All small models →