NA

Nanbeige 4.3 33B Heretic

NanbeigeGGUF by LuniUpdated 2026-10-28
Download on Hugging Face
Parameters33.1B
Context512K
Layers52
Architecturellama
Smallest quant7.7 GB
Quick answer

Nanbeige 4.3 33B Heretic needs about 22.3 GB of memory at Q4_K_M (a 20.1 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 22 quants, from 7.7 GB (IQ1_S) to 35.2 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 7.0 GB.

Nanbeige 4.3 33B Heretic – a 33B dense causal LM with IQ quants

Nanbeige 4.3 33B Heretic is a 52‑layer dense causal language model. It supports chain‑of‑thought reasoning and tools. The model is released under Apache‑2.0 and offers several IQ and Q quantizations.

Good at

  • Reasoning
  • Tools usage
  • High‑IQ STEM tasks

Limits

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

Best for

  • Academic research
  • Technical documentation generation
  • Complex problem solving

Who it’s for. Researchers and developers needing a large, reasoning‑capable language model. The model is suitable for environments that can accommodate its memory requirements and support llama.cpp inference.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S7.7 GB 9.9 GB
Too big Get file →
IQ1_M8.3 GB 10.5 GB
Too big Get file →
IQ2_XXS9.3 GB 11.6 GB
Too big Get file →
IQ2_XS10.2 GB 12.5 GB
Too big Get file →
IQ2_S10.8 GB 13.1 GB
Too big Get file →
IQ2_M11.7 GB 13.9 GB
Too big Get file →
Q2_K12.7 GB 14.9 GB
Too big Get file →
IQ3_XXS13.1 GB 15.4 GB
Too big Get file →
IQ3_XS14.0 GB 16.3 GB
Too big Get file →
Q3_K_S14.7 GB 16.9 GB
Too big Get file →
IQ3_S14.8 GB 17.0 GB
Too big Get file →
IQ3_M15.2 GB 17.5 GB
Too big Get file →
Q3_K_M16.3 GB 18.5 GB
Too big Get file →
Q3_K_L17.6 GB 19.9 GB
Too big Get file →
IQ4_XS18.0 GB 20.3 GB
Too big Get file →
IQ4_NL19.0 GB 21.2 GB
Too big Get file →
Q4_K_S19.1 GB 21.3 GB
Too big Get file →
Q4_K_M20.1 GB 22.3 GB
Too big Get file →
Q5_K_S22.9 GB 25.2 GB
Too big Get file →
Q5_K_M23.5 GB 25.7 GB
Too big Get file →
Q6_K27.2 GB 29.4 GB
Too big Get file →
Q8_035.2 GB 37.4 GB
Too big Get file →

For 8 GB VRAM use IQ1_S (file size 7.66 GB, memory 9.9 GB). For 12 GB VRAM use IQ1_M (8.28 GB, 10.5 GB). For 16 GB VRAM use IQ2_XS (10.25 GB, 12.5 GB).

What GPU do you need for Nanbeige 4.3 33B 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: 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: the best quant that fits is IQ2_XXS (9.3 GB file, about 11.6 GB in use).
  • 16 GB VRAM: the best quant that fits is IQ3_XXS (13.1 GB file, about 15.4 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_K_M (20.1 GB file, about 22.3 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (27.2 GB file, about 29.4 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (35.2 GB file, about 37.4 GB in use).

Which Nanbeige 4.3 33B Heretic quant should you choose?

Q4_K_M is the usual starting point for Nanbeige 4.3 33B Heretic: it balances file size and quality well. Here it is a 20.1 GB file.

The Q8_0 file is 35.2 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_S at 7.7 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

Architecturellama
Parameters33.1B
Layers52
Trained context524,288 tokens
KV cache (fp16)208 KB per token

Source

Quantized byLuni
FamilyNanbeige
Licenseapache-2.0
Downloads (HF)6,027
Last updated2026-10-28

Frequently asked questions about Nanbeige 4.3 33B Heretic

How much VRAM does Nanbeige 4.3 33B Heretic need?

About 22.3 GB at Q4_K_M with an 8K context window. That is the 20.1 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 Nanbeige 4.3 33B Heretic on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 9.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Nanbeige 4.3 33B Heretic GGUF?

It depends on the quant: IQ1_S is 7.7 GB, IQ1_M is 8.3 GB, IQ2_XXS is 9.3 GB, IQ2_XS is 10.2 GB, IQ2_S is 10.8 GB, IQ2_M is 11.7 GB and 16 more.

What context length does Nanbeige 4.3 33B Heretic support?

The GGUF header lists a trained context length of 524,288 tokens (512K). The KV cache grows with the context you use: about 1.7 GB at 8K and 7.0 GB at 32K.

Where can I download Nanbeige 4.3 33B Heretic GGUF?

The files are on Hugging Face in the Luni/Nanbeige-4.3-33B-Heretic-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 Nanbeige 4.3 33B 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 Nanbeige 4.3 33B Heretic on 8 GB VRAM?

Use the IQ1_S quantization; it requires approximately 9.9 GB of VRAM at 8 k context.

Does the model support image inputs?

No, the model has no vision capabilities.

What license applies to this model?

The model is released under the Apache‑2.0 license.

Similar models

Browse all →

More from Luni

All by Luni →

More Nanbeige models

All Nanbeige →

Other medium models

All medium models →