Does Qwen3.8 27B Q4_0 fit in the AMD Radeon AI PRO R9700's memory?

Estimate: the Q4_0 file is 16.1 GB and needs about 17.5 GB with an ~8K context; the AMD Radeon AI PRO R9700 has about 31.5 GB usable, so it fits.

Qwen3.8 27B › AMD Radeon AI PRO R9700

Qwen3.8 27B on the AMD Radeon AI PRO R9700

Qwen3.8 27B runs at about 34.2 tokens/s on the AMD Radeon AI PRO R9700 (measured, Q4_0, Vulkan, single-stream llama-server decode).

Best measured: 34.2 tokens/s generation (Q4_0, Vulkan). The source gives no prompt-processing figure.

For comparison, our size-scaling estimate for this combination is 36.4 tok/s.

Every run

Quant	Backend	Generate tok/s	Prompt tok/s	Test	Build / by / notes	Source
Q4_0	Vulkan	34.2	–	decode, llama-server, MTP off	Baseline (spec-off) column of a paired MTP A/B: llama-server single stream (--parallel 1), probe.py median, not llama-bench. Windows.	qwen38-mtp community table (GitHub)
UD-Q4_K_XL	Vulkan	29.7	–	decode, llama-server, MTP off	Baseline (spec-off) column of a paired MTP A/B: llama-server single stream (--parallel 1), probe.py median, not llama-bench. Windows.	qwen38-mtp community table (GitHub)
UD-Q4_K_XL	Vulkan	29.7	–	decode, llama-server, MTP off	Baseline (spec-off) column of a paired MTP A/B: llama-server single stream (--parallel 1), probe.py median, not llama-bench. Windows.	qwen38-mtp community table (GitHub)

tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128. "decode, llama-server" = single-stream generation speed measured against a running llama-server (no speculative decoding).

Check prices

Amazon Newegg

As an Amazon Associate I earn from qualifying purchases.

No card yet?

Try it on a rented GPU (RunPod)

Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.

All results for the AMD Radeon AI PRO R9700 · Qwen3.8 27B on other hardware

Embed this result

HTML

<a href="https://rigbench.pages.dev/model/qwen3-8-27b/on/amd-radeon-ai-pro-r9700/"><img src="https://rigbench.pages.dev/badge/qwen3-8-27b--amd-radeon-ai-pro-r9700.svg" alt="Qwen3.8 27B · AMD Radeon AI PRO R9700: 34.2 tok/s (Q4_0, measured)"></a>

Markdown (GitHub README, forums)

[![Qwen3.8 27B · AMD Radeon AI PRO R9700: 34.2 tok/s (Q4_0, measured)](https://rigbench.pages.dev/badge/qwen3-8-27b--amd-radeon-ai-pro-r9700.svg)](https://rigbench.pages.dev/model/qwen3-8-27b/on/amd-radeon-ai-pro-r9700/)