llamaperf

M1 Ultra 128GB vs AMD RX 7900 XTX for local LLMs

3 reports on the M1 Ultra 128GB and 32 on the AMD RX 7900 XTX

Which is faster for local LLMs?

On Qwen3.8 27B at 4-bit, the M1 Ultra 128GB runs at 25.1 tokens per second (one run) and the AMD RX 7900 XTX at 48.0 (3 runs). 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 17% higher on the AMD RX 7900 XTX (819 GB/s on the M1 Ultra 128GB, 960 on the AMD RX 7900 XTX).

M1 Ultra 128GB
Memory
128GB unified
Memory bandwidth
819 GB/s
FP16 compute
21 TFLOPS
Reports
3
AMD RX 7900 XTX
Memory
24GB
Memory bandwidth
960 GB/s
FP16 compute
123 TFLOPS
Reports
32

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

ModelM1 Ultra 128GBAMD RX 7900 XTX
Qwen3.8 27B4-bit25.1one run48.03 runs

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.

ModelM1 Ultra 128GBAMD RX 7900 XTX
Qwen3.8 27B7.5 t/sFP1634.8 t/sQ4_K_M
Qwen3.8 125B · 6B active34.0 t/sQ5_K_S19.5 t/sQ2_K, experts in RAM
DeepSeek V4 Flash 284B · 13B active53.5 t/sQ2_Kdoesn't fit at these settings
Qwen3.6 35B · 3B active23.5 t/sFP16123.2 t/sQ4_K_S
Qwen3.6 27B7.5 t/sFP1634.8 t/sQ4_K_M
GLM-5.3 320B · 18B activedoesn't fit at these settingsdoesn't fit at these settings
Gemma 4 26B · 4B active20.2 t/sFP1678.4 t/sQ5_K_S

Change the context, RAM or model in the calculator: M1 Ultra 128GB · AMD RX 7900 XTX

Every report on each card, with its source: M1 Ultra 128GB · AMD RX 7900 XTX

More comparisons