Skip to content
Run LLM Locally

Mixtral 8x7B Instruct on the RX 6800 XT

Mixtral 8x7B Instruct at Q4_K_M needs about 27.5 GB, more than the 16 GB of the RX 6800 XT: part of the model runs from system RAM, which slows generation. At Q3_K_M it does not fit in VRAM.

ShareFacebookXWhatsAppLinkedInEmail

Mixtral 8x7B Instruct

Q4_K_M · 8K context

CPU/GPU offload

The model needs about 27.5 GB but only 15.5 GB of VRAM is usable: 11.9 GB must be loaded into system RAM. It runs, 57% on the GPU, but more slowly.

Your hardware
RX 6800 XT 16 GB · 64 GB RAM
Estimated memory
~27.5 GB estimate
VRAM (usable memory)
15.5 GB
GPU only
No
CPU/GPU offload
Required (11.9 GB)
Generation speed
5–10 tok/s estimate
Largest context in fast memory
None
Confidence: low

Based on: memory bandwidth (512 GB/s) and model size · system RAM bandwidth (~51 GB/s)

Memory breakdown

~27.5 GB (26.7 GB–28.2 GB) needed · estimate

11.9 GB must be offloaded to system RAM

  • Model weights24.6 GB
  • KV cache1 GB
  • Runtime overhead0.5 GB
  • Safety margin1.3 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
  • split CPU/GPU runs vary a lot with the CPU and PCIe link
  • system RAM speed unknown (dual-channel DDR4-3200 assumed)
  • mixture-of-experts: speed depends on expert routing

Other quantizations on the RX 6800 XT

Mixtral 8x7B Instruct on RX 6800 XT + 64 GB RAM · 8K context
Mixtral 8x7B Instruct Q8_08K~50.1 GB · 1.9–3.5 tok/s estimatePossible but slow
Mixtral 8x7B Instruct Q6_K8K~39.1 GB · 2.5–5.5 tok/s estimatePossible but slow
Mixtral 8x7B Instruct Q5_K_M8K~33.1 GB · 3.5–7 tok/s estimatePossible but slow
Mixtral 8x7B Instruct Q4_K_M8K~27.5 GB · 5–10 tok/s estimateCPU/GPU offload
Mixtral 8x7B Instruct IQ4_XS8K~27.1 GB · 5–11 tok/s estimateCPU/GPU offload
Mixtral 8x7B Instruct Q3_K_M8K~21.5 GB · 9–18 tok/s estimateCPU/GPU offload

Alternatives

GPUs that run Mixtral 8x7B Instruct Q4_K_M entirely in VRAM

Other models that fit the RX 6800 XT

Hardware that runs Mixtral 8x7B Instruct Q4_K_M entirely in memory

Prices and links follow the market you choose.

GeForce RTX 5090

NVIDIA

CompatibleRuns Mixtral 8x7B Instruct Q4_K_M entirely in memory

Price not available

Availability unknown

View on amazon.com

Affiliate link

Radeon PRO W7900

AMD

ExcellentRuns Mixtral 8x7B Instruct Q4_K_M entirely in memory

Price not available

Availability unknown

View on amazon.com

Affiliate link

MacBook Pro 14 M3 Max

Apple Silicon

ExcellentRuns Mixtral 8x7B Instruct Q4_K_M entirely in memory

Price not available

Availability unknown

View on amazon.com

Affiliate link

AMD Ryzen AI Max+ 395

Unified memory PC

ExcellentRuns Mixtral 8x7B Instruct 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 6800 XT run Mixtral 8x7B Instruct?
The model needs about 27.5 GB but only 15.5 GB of VRAM is usable: 11.9 GB must be loaded into system RAM. It runs, 57% on the GPU, but more slowly.
Which quantization should I use on the RX 6800 XT?
None fits entirely in VRAM at 8K context; the smallest, Q3_K_M, still needs offload.
What is the largest context on the RX 6800 XT?
At Q4_K_M, the model does not fit entirely in VRAM even at a short context.
How fast is Mixtral 8x7B Instruct (RX 6800 XT)?
We estimate 5–10 tok/s (estimate, confidence low). Based on: memory bandwidth (512 GB/s) and model size, system RAM bandwidth (~51 GB/s).