QW

Qwen3.6 35B A3B REAM 160 ru agent

QwenGGUF by WaveCutUpdated 2026-06-21
Download on Hugging Face
Parameters23.1B
Context256K
Layers41
Architectureqwen35moe
Smallest quant4.2 GB
Quick answer

Qwen3.6 35B A3B REAM 160 ru agent needs about 15.6 GB of memory at Q4_K_M (a 14.4 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 34 quants, from 4.2 GB (Q1_0) to 46.3 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 2.8 GB.

Qwen3.6 35B A3B REAM 160 ru agent – a 23.1B‑parameter MoE model compressed with REAM

This model is a 23.1‑billion‑parameter Qwen architecture with 160 Mixture‑of‑Experts layers. It is compressed using REAM and fine‑tuned for agentic and Russian language tasks. The GGUF quantizations are available for a range of memory budgets.

Good at

  • Agentic tool‑use
  • Russian Cyrillic chat
  • Code and math reasoning

Limits

  • No vision support
  • Large context requires high VRAM
  • License not specified

Best for

  • Developing Russian‑language agents
  • Coding assistance in Russian
  • Tool‑calling workflows

Who it’s for. Engineers and researchers building Russian‑language agents or code‑generation tools. The model is suitable for those who need a large MoE model with specialized calibration for Russian and agentic tasks.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q1_04.2 GB 5.4 GB
Fits Get file →
IQ1_S5.6 GB 6.8 GB
Fits Get file →
IQ1_M6.1 GB 7.3 GB
Tight Get file →
IQ2_XXS6.9 GB 8.1 GB
Too big Get file →
IQ2_XS7.5 GB 8.7 GB
Too big Get file →
IQ2_S7.7 GB 8.9 GB
Too big Get file →
IQ2_M8.3 GB 9.5 GB
Too big Get file →
Q2_K_S8.6 GB 9.8 GB
Too big Get file →
Q2_K9.1 GB 10.3 GB
Too big Get file →
IQ3_XXS9.5 GB 10.7 GB
Too big Get file →
IQ3_XS10.1 GB 11.3 GB
Too big Get file →
Q3_K_S10.5 GB 11.7 GB
Too big Get file →
IQ3_S10.6 GB 11.8 GB
Too big Get file →
IQ3_M10.7 GB 11.9 GB
Too big Get file →
Q3_K_M11.6 GB 12.7 GB
Too big Get file →
Q3_K_XL12.1 GB 13.3 GB
Too big Get file →
Q3_K_L12.4 GB 13.6 GB
Too big Get file →
IQ4_XS12.9 GB 14.0 GB
Too big Get file →
Q4_013.5 GB 14.7 GB
Too big Get file →
IQ4_NL13.5 GB 14.7 GB
Too big Get file →
Q4_K_S13.6 GB 14.8 GB
Too big Get file →
MXFP4_MOE13.9 GB 15.1 GB
Too big Get file →
Q4_K_M14.4 GB 15.6 GB
Too big Get file →
Q4_114.8 GB 16.0 GB
Too big Get file →
Q4_K_XL14.9 GB 16.0 GB
Too big Get file →
Q5_K_S16.2 GB 17.4 GB
Too big Get file →
Q5_016.3 GB 17.5 GB
Too big Get file →
Q5_K_M16.7 GB 17.9 GB
Too big Get file →
Q5_K_XL17.1 GB 18.3 GB
Too big Get file →
Q5_117.6 GB 18.8 GB
Too big Get file →
Q6_K19.2 GB 20.3 GB
Too big Get file →
Q6_K_XL19.4 GB 20.6 GB
Too big Get file →
Q8_024.6 GB 25.8 GB
Too big Get file →
BF1646.3 GB 47.5 GB
Too big Get file →

For 8 GB VRAM use Q1_0 (4.16 GB file, 5.4 GB memory). For 12 GB VRAM use Q4_K_S (13.58 GB file, 14.8 GB memory). For 20 GB VRAM use Q6_K (19.15 GB file, 20.3 GB memory). Estimates are approximate.

What GPU do you need for Qwen3.6 35B A3B REAM 160 ru agent?

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 Q1_0 (4.2 GB file, about 5.4 GB in use).
  • 8 GB VRAM: the best quant that fits is IQ1_M (6.1 GB file, about 7.3 GB in use).
  • 12 GB VRAM: the best quant that fits is IQ3_M (10.7 GB file, about 11.9 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_XL (14.9 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K_XL (19.4 GB file, about 20.6 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (24.6 GB file, about 25.8 GB in use).
  • 48 GB VRAM: the best quant that fits is BF16 (46.3 GB file, about 47.5 GB in use).

Which Qwen3.6 35B A3B REAM 160 ru agent quant should you choose?

Q4_K_M is the usual starting point for Qwen3.6 35B A3B REAM 160 ru agent: it balances file size and quality well. Here it is a 14.4 GB file.

The Q8_0 file is 24.6 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 Q1_0 at 4.2 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35moe
Parameters23.1B (approximate, mixture of experts)
Active parameters3B
Layers41
Trained context262,144 tokens
KV cache (fp16)82 KB per token

Source

Quantized byWaveCut
FamilyQwen
Downloads (HF)2,673
Last updated2026-06-21

Frequently asked questions about Qwen3.6 35B A3B REAM 160 ru agent

How much VRAM does Qwen3.6 35B A3B REAM 160 ru agent need?

About 15.6 GB at Q4_K_M with an 8K context window. That is the 14.4 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 Qwen3.6 35B A3B REAM 160 ru agent on an 8 GB GPU?

Yes, with IQ1_M: the 6.1 GB file needs about 7.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Qwen3.6 35B A3B REAM 160 ru agent GGUF?

It depends on the quant: Q1_0 is 4.2 GB, IQ1_S is 5.6 GB, IQ1_M is 6.1 GB, IQ2_XXS is 6.9 GB, IQ2_XS is 7.5 GB, IQ2_S is 7.7 GB and 28 more.

What context length does Qwen3.6 35B A3B REAM 160 ru agent support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.8 GB at 32K.

Where can I download Qwen3.6 35B A3B REAM 160 ru agent GGUF?

The files are on Hugging Face in the WaveCut/Qwen3.6-35B-A3B-REAM-160-ru-agent-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run this model on 8 GB VRAM?

Yes, the Q1_0 quantization requires about 5.4 GB of VRAM at 8 k context.

Does this model support vision?

No, the model has no vision capabilities.

What is the context length?

The model was trained with 262,144 context tokens.

Other GGUFs of Qwen3.6-35B-A3B-REAM-160-ru-agent

Similar models

Browse all →

More from WaveCut

All by WaveCut →

More Qwen models

All Qwen →

Other medium models

All medium models →