GI

GigaChat 20B A3B Instruct

GigachatGGUF by mradermacherUpdated 2025-01-11
Download on Hugging Face
Parameters20.6B
Context128K
Layers28
Architecturedeepseek
Smallest quant4.3 GB
Quick answer

GigaChat 20B A3B Instruct needs about 13.9 GB of memory at Q4_K_M (a 12.5 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 23 quants, from 4.3 GB (IQ1_S) to 16.9 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.

GigaChat 20B A3B Instruct is a 20.6B-parameter Gigachat model in GGUF format, quantized by mradermacher. Its trained context length is 128K tokens. The repository has 23 quant files, from 4.3 GB (IQ1_S) to 16.9 GB (Q6_K).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S4.3 GB 5.7 GB
Fits Get file →
IQ1_M4.7 GB 6.2 GB
Fits Get file →
IQ2_XXS5.5 GB 6.9 GB
Fits Get file →
IQ2_XS6.1 GB 7.6 GB
Tight Get file →
IQ2_S6.2 GB 7.6 GB
Tight Get file →
IQ2_M6.8 GB 8.2 GB
Too big Get file →
Q2_K_S7.0 GB 8.5 GB
Too big Get file →
Q2_K7.6 GB 9.0 GB
Too big Get file →
IQ3_XXS8.0 GB 9.4 GB
Too big Get file →
IQ3_XS8.5 GB 9.9 GB
Too big Get file →
IQ3_S9.0 GB 10.4 GB
Too big Get file →
Q3_K_S9.0 GB 10.4 GB
Too big Get file →
IQ3_M9.1 GB 10.5 GB
Too big Get file →
Q3_K_M9.9 GB 11.3 GB
Too big Get file →
Q3_K_L10.7 GB 12.1 GB
Too big Get file →
IQ4_XS11.0 GB 12.5 GB
Too big Get file →
Q4_011.7 GB 13.1 GB
Too big Get file →
Q4_K_S11.7 GB 13.2 GB
Too big Get file →
Q4_K_M12.5 GB 13.9 GB
Too big Get file →
Q4_112.9 GB 14.4 GB
Too big Get file →
Q5_K_S14.2 GB 15.6 GB
Too big Get file →
Q5_K_M14.6 GB 16.1 GB
Too big Get file →
Q6_K16.9 GB 18.3 GB
Too big Get file →

What GPU do you need for GigaChat 20B A3B 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: no quant fits at an 8K context.
  • 6 GB VRAM: the best quant that fits is IQ1_S (4.3 GB file, about 5.7 GB in use).
  • 8 GB VRAM: the best quant that fits is IQ2_S (6.2 GB file, about 7.6 GB in use).
  • 12 GB VRAM: the best quant that fits is Q3_K_M (9.9 GB file, about 11.3 GB in use).
  • 16 GB VRAM: the best quant that fits is Q5_K_S (14.2 GB file, about 15.6 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (16.9 GB file, about 18.3 GB in use).

Which GigaChat 20B A3B Instruct quant should you choose?

Q4_K_M is the usual starting point for GigaChat 20B A3B Instruct: it balances file size and quality well. Here it is a 12.5 GB file.

If memory is tight, the smallest file is IQ1_S at 4.3 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

Architecturedeepseek
Parameters20.6B (approximate, mixture of experts)
Active parameters3B
Layers28
Trained context131,072 tokens
KV cache (fp16)112 KB per token

Source

Quantized bymradermacher
FamilyGigachat
Licensemit
Downloads (HF)304
Last updated2025-01-11

Frequently asked questions about GigaChat 20B A3B Instruct

How much VRAM does GigaChat 20B A3B Instruct need?

About 13.9 GB at Q4_K_M with an 8K context window. That is the 12.5 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 GigaChat 20B A3B Instruct on an 8 GB GPU?

Yes, with IQ2_S: the 6.2 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 GigaChat 20B A3B Instruct GGUF?

It depends on the quant: IQ1_S is 4.3 GB, IQ1_M is 4.7 GB, IQ2_XXS is 5.5 GB, IQ2_XS is 6.1 GB, IQ2_S is 6.2 GB, IQ2_M is 6.8 GB and 17 more.

What context length does GigaChat 20B A3B Instruct support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.

Where can I download GigaChat 20B A3B Instruct GGUF?

The files are on Hugging Face in the mradermacher/GigaChat-20B-A3B-instruct-i1-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 GigaChat 20B A3B Instruct use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Other GGUFs of GigaChat-20B-A3B-instruct

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Gigachat models

All Gigachat →

Other medium models

All medium models →