RunsLocalGPU checkerLocal AI toolsv1.8 · data Oct 14, 2026

All models / Qwen3.6 35B-A3B

What GPU do I need to run Qwen3.6 35B-A3B?

Qwen3.6 35B-A3B has 35B parameters (3B active per token, mixture-of-experts). At Q4_K_M with 8K context it needs about 22.2 GB of memory. The smallest graphics card that runs it fully on the GPU is about 24 GB, such as the RTX 3090. Because only 3B parameters are active, it still runs usably with offload to system RAM on cards like the RX 7900 XT.

Check your own hardware

Memory by quantization (8K context)

Hardware that runs it fully in fast memory

Runs with offload to system RAM (slower)

RX 7900 XT ~114 tok/s · RTX 5080 ~65 tok/s · RTX 5070 Ti ~64 tok/s · RTX 4080 Super ~62 tok/s · RTX 4080 ~62 tok/s · RTX 4070 Ti Super ~61 tok/s · RX 9070 XT ~60 tok/s · RX 9070 ~60 tok/s · RX 7800 XT ~60 tok/s · Arc A770 16GB ~59 tok/s · RX 6800 XT ~57 tok/s · RTX 5060 Ti 16GB ~56 tok/s · RX 9060 XT 16GB ~50 tok/s · RTX 4060 Ti 16GB ~49 tok/s · RTX 3080 Ti ~44 tok/s · RTX 3080 12GB ~44 tok/s · RTX 5070 ~43 tok/s · RTX 4070 Ti ~41 tok/s · RTX 4070 Super ~41 tok/s · RTX 4070 ~41 tok/s · Arc B580 ~41 tok/s · RX 7700 XT ~41 tok/s · RTX 2080 Ti ~39 tok/s · RTX 3060 12GB ~39 tok/s

Assumes 32 GB of DDR5 system RAM for offload. Estimates only; real speeds usually land within about 20%.