DeepSeek V3.1 (Q2_K_XL) — 127.4 GBon AMD Radeon 8060S 128GB

Q2_K_XL AMD Radeon 8060S 128GB

Overview

DeepSeek V3.1 is a 684.53B parameter moe language model by DeepSeek, with code, multilingual, thinking, tool-calls capabilities. It supports a context window of up to 163,840 tokens.

DeepSeek V3.1 is a 685-billion-parameter Mixture-of-Experts model from DeepSeek, activating 8 of 256 experts per token plus one shared expert. It delivers frontier-level performance on code generation, reasoning, and multilingual tasks while using far fewer active parameters per inference step than comparably sized dense models. The model supports thinking mode, tool calling, and nine languages. With a 160K context window, it requires multi-GPU or distributed setups but quantizes down to Q2 levels for reduced VRAM footprint.

At Q2_K_XL quantization (low quality tier), the model weighs 238.17 GB. This exceeds the 128 GB of VRAM on AMD Radeon 8060S 128GB. Inference is still possible via CPU offload or memory-mapped loading from disk, but expect significantly reduced performance.

Hardware Requirements

Model size	238.17 GB
VRAM available	128 GB
VRAM used	127.4 GB
Min RAM required	233.9 GB
GPU layers	61 / 61
Context size	74,831
Backend	vulkan
Flash attention	Yes

Performance Notes

Mixture-of-Experts

This is a Mixture-of-Experts model with 256 total experts, of which 8 are active per token. Only the active experts need fast GPU access — inactive expert weights can reside in system RAM with minimal impact on generation speed.

Read the MoE deployment guide →

Possible performance impact

Expert weights offloaded to CPU via tensor override. At least 233.9 GB system RAM recommended — if insufficient, data will be read from disk on demand.

Shorter conversations

Context size has been reduced due to memory constraints. This may affect the model's ability to handle long conversations.

Deploy

Server Local

Command

helmfile --state-values-file <(curl -s https://www.prositronic.eu/values/deepseek-v3-1/q2_k_xl/amd-8060s-128gb.yaml) apply

Generated values.yaml

/values/deepseek-v3-1/q2_k_xl/amd-8060s-128gb.yaml

Loading values…

Install llama.cpp

Install llama.cpp from the official build instructions for your platform and backend.

Download Model

curl -L -o deepseek-v3-1.gguf "https://huggingface.co/unsloth/DeepSeek-V3.1-GGUF/resolve/main/UD-Q2_K_XL/DeepSeek-V3.1-UD-Q2_K_XL-00001-of-00006.gguf"

Start Server

llama-server \
  -m deepseek-v3-1.gguf \
  --n-gpu-layers 61 \
  --ctx-size 74831 \
  --flash-attn \
  -ot ".ffn_.*_exps.=CPU"

Verify

curl http://localhost:8080/health

Explore

• Configure a deployment
• Browse models
• Browse hardware

Other quantizations

• DeepSeek V3.1 (Q8_0)
• DeepSeek V3.1 (Q5_K_M)
• DeepSeek V3.1 (Q4_K_M)

Other hardware

• DeepSeek V3.1 on CPU Only
• DeepSeek V3.1 on NVIDIA A100 80GB
• DeepSeek V3.1 on NVIDIA H100

Frequently Asked Questions

How much VRAM does DeepSeek V3.1 (Q2_K_XL) need?

The Q2_K_XL quantization of DeepSeek V3.1 requires 238.17 GB. All 61 layers fit in the 128 GB of VRAM available on AMD Radeon 8060S 128GB, enabling full GPU acceleration.

Can I run DeepSeek V3.1 on AMD Radeon 8060S 128GB?

Yes. AMD Radeon 8060S 128GB provides 128 GB of VRAM, which is enough to run DeepSeek V3.1 (Q2_K_XL) with all layers on the GPU for optimal performance.

What is quantization?

Quantization reduces a model's numerical precision from its original floating-point format to a more compact representation. This shrinks the file size and VRAM footprint, making it possible to run large models on consumer hardware. The trade-off is a small reduction in output quality. Q2_K_XL compresses DeepSeek V3.1 from its original size down to 238.17 GB.

What quantization should I choose for DeepSeek V3.1?

Q2_K_XL is a low-quality quantization. Higher-quality quants (Q8, Q6) preserve more model accuracy but need more VRAM. Lower quants (Q4, Q3, Q2) reduce VRAM usage at the cost of some quality. Choose based on your available hardware and quality requirements.

What is flash attention and why is it enabled?

Flash attention is a memory-efficient algorithm that speeds up the attention mechanism in transformer models. It reduces VRAM usage during inference by avoiding the materialisation of the full attention matrix. For DeepSeek V3.1 on AMD Radeon 8060S 128GB, flash attention is enabled to maximise context length and throughput within the available 128 GB of VRAM.

What is MoE and how does it affect deployment?

DeepSeek V3.1 uses a Mixture-of-Experts (MoE) architecture with 256 experts, of which 8 are active per token. This means only a fraction of the model weights are used for each inference step, allowing MoE models to be larger in total parameter count while remaining efficient at inference time.

Last updated: March 5, 2026