How fast does Qwen3.6 35B-A3B run on the AMD Radeon 780M?

Qwen3.6 35B-A3B runs at about 20.5 tokens/s on the AMD Radeon 780M (measured, Q8_0, Vulkan, llama-bench). Based on 2 measured run(s), each linked to its public source.

Qwen3.6 35B-A3B › AMD Radeon 780M

Qwen3.6 35B-A3B on the AMD Radeon 780M

Qwen3.6 35B-A3B runs at about 20.5 tokens/s on the AMD Radeon 780M (measured, Q8_0, Vulkan, llama-bench).

Best measured: 20.5 tokens/s generation and 325 tokens/s prompt processing (Q8_0, Vulkan, FA on).

For comparison, our size-scaling estimate for this combination is 11.4 tok/s.

Every run

Quant	Backend	Generate tok/s	Prompt tok/s	Test	Build / by / notes	Source
Q8_0	Vulkan FA on	20.5	325	tg128	c841aee MINISFORUM UM790 Pro (Radeon 780M iGPU); ggml-org Q8_0 GGUF.	llama.cpp discussion #27519
Q8_0	ROCm FA on	15.8	338	tg128	c841aee MINISFORUM UM790 Pro (Radeon 780M iGPU); ggml-org Q8_0 GGUF.	llama.cpp discussion #27519

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 780M · Qwen3.6 35B-A3B on other hardware

Embed this result

HTML

<a href="https://rigbench.pages.dev/model/qwen3-6-35b-a3b/on/amd-radeon-780m/"><img src="https://rigbench.pages.dev/badge/qwen3-6-35b-a3b--amd-radeon-780m.svg" alt="Qwen3.6 35B-A3B · AMD Radeon 780M: 20.5 tok/s (Q8_0, measured)"></a>

Markdown (GitHub README, forums)

[![Qwen3.6 35B-A3B · AMD Radeon 780M: 20.5 tok/s (Q8_0, measured)](https://rigbench.pages.dev/badge/qwen3-6-35b-a3b--amd-radeon-780m.svg)](https://rigbench.pages.dev/model/qwen3-6-35b-a3b/on/amd-radeon-780m/)