phi-4

Phi family · 15.0B params

Evidence: 1 sourced run · 1 independent source · first recorded Aug 18, 2026. What these labels mean.

What the phi-4 numbers mean

The fastest verified result on our board is 34.2 output tok/s on an RTX 3060 x2 at Q4_K_M with a 16384-token context. That is about 29.2 ms per generated token — 500 tokens takes roughly 14.6 seconds. Prompt processing on the same record runs at 889.5 tok/s, so a 2,000-token prompt is read in about 2.2 seconds. Peak VRAM was measured at 9.8 GB.

How much VRAM does phi-4 need?

phi-4 has 15.0B parameters. These estimates use the same formula as the VRAM calculator: weights, plus about 15% for a moderate-context KV cache and runtime overhead, plus a small engine floor.

Long contexts and large batches raise the real number — read the smallest card as "possible with a reduced context". Measured throughput is on the benchmark board.

Models with a similar footprint

Ranked by distance from phi-4's parameter count. Full benchmark board.

All benchmarks Methodology