How much VRAM does the AMD Radeon RX 6800 XT have?

The AMD Radeon RX 6800 XT has 16 GB of GDDR6 with 512 GB/s memory bandwidth.

What is the AMD Radeon RX 6800 XT best for?

With 16 GB of VRAM, the AMD Radeon RX 6800 XT handles smaller models (7B–14B) at Q4–Q5 quantization — ideal for entry-level local LLM experimentation and lightweight inference.

What LLMs can the AMD Radeon RX 6800 XT run locally?

The AMD Radeon RX 6800 XT can run 35 of the 80 open-weight models tracked by CanItRun natively in VRAM at 8k context. Top options include: Llama 3.1 8B Instruct at Q8_0, Llama 3.2 3B Instruct at BF16, Llama 3.2 1B Instruct at FP32.

Can the AMD Radeon RX 6800 XT run Llama 3.3 70B Instruct?

The AMD Radeon RX 6800 XT can run Llama 3.3 70B Instruct with CPU offload at Q3_K_M quantization, but inference will be slower than native VRAM execution.

Can the AMD Radeon RX 6800 XT run Qwen 3.6 27B?

Yes. The AMD Radeon RX 6800 XT runs Qwen 3.6 27B natively in VRAM at Q2_K quantization, achieving approximately 28 tokens per second.

Can the AMD Radeon RX 6800 XT run Llama 3.1 8B Instruct?

Yes. The AMD Radeon RX 6800 XT runs Llama 3.1 8B Instruct natively in VRAM at Q8_0 quantization, achieving approximately 34.7 tokens per second.

AMD Radeon RX 6800 XT

The AMD Radeon RX 6800 XT has 16 GB VRAM and 512 GB/s memory bandwidth. It can run 35 of our 80 tracked models natively in VRAM at 8k context.

With 16 GB GDDR6, the AMD Radeon RX 6800 XT is a consumer-tier GPU that can run 35 models natively. It handles 30B-class models at Q4 quantization.

AMD Radeon RX 6800 XT: 2020 RDNA 2 with 16GB GDDR6 at 512 GB/s — used market value.

7B-14B at Q4 native. ~6-10 t/s for 7B.

ROCm Linux support is maturing. Vulkan works cross-platform. Best budget AMD option.

Vendor	AMD
Architecture	RDNA 2
VRAM	16 GB
Memory type	GDDR6
Memory bandwidth	512 GB/s
Compute backend	ROCM
Tier	Consumer
Released	2020
Models (native)	35 / 80
Models (offload)	14 / 80

Software: ROCm is Linux-only; on Windows use the Vulkan backend instead. Requires llama.cpp compiled with ROCm support.

Popular models for this GPU

Nemotron 3 Nano 30B Qwen3 30B-A3B (MoE)Gemma 2 27B Instruct Gemma 3 27B Instruct Qwen 3.6 27B

Models this GPU runs natively in VRAM (35)

Models that fit with CPU offload (14)

These use system RAM for layers that don't fit in VRAM — expect much slower inference.

Too large for this GPU (31)

Compare AMD Radeon RX 6800 XT with other GPUs

AMD Radeon RX 6800 XTvsNVIDIA RTX 408016 GB each

Continue reading

hardware10 min

AMD Radeon for LLMs: ROCm & Vulkan Complete Guide

Frequently asked questions

How much VRAM does the AMD Radeon RX 6800 XT have?: The AMD Radeon RX 6800 XT has 16 GB of GDDR6 with 512 GB/s memory bandwidth.
What is the AMD Radeon RX 6800 XT best for?: With 16 GB of VRAM, the AMD Radeon RX 6800 XT handles smaller models (7B–14B) at Q4–Q5 quantization — ideal for entry-level local LLM experimentation and lightweight inference.
What LLMs can the AMD Radeon RX 6800 XT run locally?: The AMD Radeon RX 6800 XT can run 35 of the 80 open-weight models tracked by CanItRun natively in VRAM at 8k context. Top options include: Llama 3.1 8B Instruct at Q8_0, Llama 3.2 3B Instruct at BF16, Llama 3.2 1B Instruct at FP32.
Can the AMD Radeon RX 6800 XT run Llama 3.3 70B Instruct?: The AMD Radeon RX 6800 XT can run Llama 3.3 70B Instruct with CPU offload at Q3_K_M quantization, but inference will be slower than native VRAM execution.
Can the AMD Radeon RX 6800 XT run Qwen 3.6 27B?: Yes. The AMD Radeon RX 6800 XT runs Qwen 3.6 27B natively in VRAM at Q2_K quantization, achieving approximately 28 tokens per second.
Can the AMD Radeon RX 6800 XT run Llama 3.1 8B Instruct?: Yes. The AMD Radeon RX 6800 XT runs Llama 3.1 8B Instruct natively in VRAM at Q8_0 quantization, achieving approximately 34.7 tokens per second.