llamaperf

M5 Ultra 256GB vs AMD RX 7800 XT 16GB for local LLMs

12 reports on the M5 Ultra 256GB and 3 on the AMD RX 7800 XT 16GB

Which is faster for local LLMs?

On Qwen3.8 27B at 4-bit, the M5 Ultra 256GB runs at 50.0 tokens per second (one run) and the AMD RX 7800 XT 16GB at 30.0 (one run). Those are community reports whose engines and settings differ, so the gap is a rough sign, not a controlled measure of the two cards. Memory bandwidth, which caps how fast a card can write, is 2.0x higher on the M5 Ultra 256GB (1,229 GB/s on the M5 Ultra 256GB, 624 on the AMD RX 7800 XT 16GB).

M5 Ultra 256GB
Memory
256GB unified
Memory bandwidth
1,229 GB/s
FP16 compute
not on record
Reports
12
AMD RX 7800 XT 16GB
Memory
16GB
Memory bandwidth
624 GB/s
FP16 compute
75 TFLOPS
Reports
3

Measured on both

Median tokens per second from plain runs (one device, one request, no speculative decoding, the whole model in the device's memory) of the same model size at the same quant level. Engines and context lengths can still differ between the runs. How to read these

ModelM5 Ultra 256GBAMD RX 7800 XT 16GB
Qwen3.8 27B4-bit50.0one run30.0one run

Estimated by the calculator

The quant the calculator recommends for each card and its estimated speed at a 32,768-token context with 32 GB of system RAM, from memory bandwidth and the model's shape. Where a model was also measured above, the measurement wins.

ModelM5 Ultra 256GBAMD RX 7800 XT 16GB
Qwen3.8 27B14.9 t/sFP1640.7 t/sQ2_K
Qwen3.8 125B · 6B active50.8 t/sQ8_0doesn't fit at these settings
DeepSeek V4 Flash 284B · 13B active66.9 t/sQ4_K_Mdoesn't fit at these settings
Qwen3.6 35B · 3B active53.5 t/sFP16134.6 t/sQ2_K
Qwen3.6 27B14.9 t/sFP1640.7 t/sQ2_K
Gemma 4 26B · 4B active40.1 t/sFP1690.7 t/sQ3_K_S
DeepSeek V4.1 Flash 552B · 16B active89.4 t/sQ2_Kdoesn't fit at these settings

Change the context, RAM or model in the calculator: M5 Ultra 256GB · AMD RX 7800 XT 16GB

Every report on each card, with its source: M5 Ultra 256GB · AMD RX 7800 XT 16GB

More comparisons