Skip to content
Run LLM Locally

Qwen3 14B on the RX 7800 XT

With 16 GB of VRAM, the RX 7800 XT runs Qwen3 14B at Q4_K_M entirely on the GPU with an 8K context (about 10.9 GB needed). At Q8_0 it does not fit in VRAM.

ShareFacebookXWhatsAppLinkedInEmail

Qwen3 14B

Q4_K_M · 8K context

Excellent

The model needs about 10.9 GB and fits in the 15.5 GB available, with 4.6 GB to spare.

Your hardware
RX 7800 XT 16 GB · 64 GB RAM
Estimated memory
~10.9 GB estimate
VRAM (usable memory)
15.5 GB
GPU only
Yes
CPU/GPU offload
Not required
Generation speed
38–60 tok/s estimate
Largest context in fast memory
36K tokens
Confidence: low

Based on: memory bandwidth (624 GB/s) and model size

Memory breakdown

~10.9 GB (10.4 GB–11.5 GB) needed · estimate

Fits entirely in VRAM · 4.6 GB left

  • Model weights8.4 GB
  • KV cache1.3 GB
  • Runtime overhead0.8 GB
  • Safety margin0.5 GB
Assumptions: context 8K tokens · KV cache f16 · llama.cpp (Vulkan) · flash attention on · weights = real file size

EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card

What makes this estimate less reliable

  • no benchmark for this hardware and model yet

Other quantizations on the RX 7800 XT

Qwen3 14B on RX 7800 XT + 64 GB RAM · 8K context
Qwen3 14B Q8_08K~17.5 GB · 7–14 tok/s estimateCPU/GPU offload
Qwen3 14B Q6_K8K~14 GB · 29–45 tok/s estimateCompatible
Qwen3 14B Q5_K_M8K~12.4 GB · 33–52 tok/s estimateExcellent
Qwen3 14B Q4_K_M8K~10.9 GB · 38–60 tok/s estimateExcellent
Qwen3 14B IQ4_XS8K~10.1 GB · 41–65 tok/s estimateExcellent
Qwen3 14B Q3_K_M8K~9.3 GB · 45–72 tok/s estimateExcellent

Alternatives

GPUs that run Qwen3 14B Q4_K_M entirely in VRAM

Other models that fit the RX 7800 XT

Where to buy a RX 7800 XT

Prices and links follow the market you choose.

Radeon RX 7800 XT

16 GB GDDR6 · 624 GB/s · 263 W

ExcellentRuns Qwen3 14B Q4_K_M entirely in memory

Price not available

Availability unknown

View on amazon.com

Affiliate link

No current offer imported for this market: the link opens a marketplace search.

Frequently asked questions

Can the RX 7800 XT run Qwen3 14B?
The model needs about 10.9 GB and fits in the 15.5 GB available, with 4.6 GB to spare.
Which quantization should I use on the RX 7800 XT?
Q6_K is the highest-precision quantization that fits entirely in VRAM at 8K context. ~14 GB / 15.5 GB.
What is the largest context on the RX 7800 XT?
At Q4_K_M, about 36K tokens fit entirely in VRAM with an f16 KV cache. A q8_0 KV cache roughly doubles that.
How fast is Qwen3 14B (RX 7800 XT)?
We estimate 38–60 tok/s (estimate, confidence low). Based on: memory bandwidth (624 GB/s) and model size.