llamaperf

Intel Arc Pro B60 24GB vs M3 Max 128GB for local LLMs

5 reports on the Intel Arc Pro B60 24GB and 6 on the M3 Max 128GB

Which is faster for local LLMs?

On Qwen3.5 35B · 3B active at 4-bit, the Intel Arc Pro B60 24GB runs at 8.3 tokens per second (one run) and the M3 Max 128GB at 90.8 (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 11% higher on the Intel Arc Pro B60 24GB (456 GB/s on the Intel Arc Pro B60 24GB, 410 on the M3 Max 128GB).

Intel Arc Pro B60 24GB
Memory
24GB
Memory bandwidth
456 GB/s
FP16 compute
98.5 TFLOPS
Reports
5
M3 Max 128GB
Memory
128GB unified
Memory bandwidth
410 GB/s
FP16 compute
14.2 TFLOPS
Reports
6

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

ModelIntel Arc Pro B60 24GBM3 Max 128GB
Qwen3.5 35B · 3B active4-bit8.3one run90.8one 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.

ModelIntel Arc Pro B60 24GBM3 Max 128GB
Qwen3.5 35B · 3B active58.5 t/sQ4_K_S16.8 t/sFP16
Qwen3.8 27B16.5 t/sQ4_K_M5.3 t/sFP16
Qwen3.8 125B · 6B active17.8 t/sQ2_K, experts in RAM24.3 t/sQ5_K_S
DeepSeek V4 Flash 284B · 13B activedoesn't fit at these settings38.2 t/sQ2_K
Qwen3.6 35B · 3B active58.5 t/sQ4_K_S16.8 t/sFP16
Qwen3.6 27B16.5 t/sQ4_K_M5.3 t/sFP16
Gemma 4 26B · 4B active37.2 t/sQ5_K_S14.4 t/sFP16

Change the context, RAM or model in the calculator: Intel Arc Pro B60 24GB · M3 Max 128GB

Every report on each card, with its source: Intel Arc Pro B60 24GB · M3 Max 128GB

More comparisons