Neural Inference | AI & Hardware

The ultimate mobile workstation. With 64GB of unified memory and the M4 Max chip, this machine targets the sweet spot between portability and extreme inference capability, enabling larger model quantization without leaving the backpack.

GPUApple M4 Max

VRAMShared

CPUM4 Max

RAM64GB Unified

Storage2TB NVMe SSD

OSmacOS Sequoia

Shop This Rig ↗

Performance Log

Date	Model	Hardware	Params	Context	Prompt Eval	Token Gen
2026-01-29	Gemma-3-4.3B	Apple M4 Max	Q4_K_M	7333 tk	1540.5 t/s	108.4 t/s
2026-01-29	Gemma-3-4.3B (32k Context)	Apple M4 Max	Q4_K_M	32768 tk	1602.0 t/s	103.9 t/s
2026-01-29	Gemma-3-4.3B (64k Context)	Apple M4 Max	Q4_K_M	65536 tk	1377.8 t/s	96.2 t/s

SPONSORED TESTS// AD_SLOT: 1234567890 // FORMAT: AUTO

Performance Analysis

Trend Visualization

Historical Archive

Filter:

Showing 3 of 3 runs

Date	Model	Quant	Context	VRAM	Prompt (t/s)	Gen (t/s)
2026-01-29	Gemma-3-4.3B	Q4_K_M	7,333	-	1540.5	108.4
2026-01-29	Gemma-3-4.3B (32k Context)	Q4_K_M	32,768	-	1602.0	103.9
2026-01-29	Gemma-3-4.3B (64k Context)	Q4_K_M	65,536	-	1377.8	96.2