Qwen 3.8 Flash

Qwen family · 125.0B params

Evidence: 8 sourced runs · 1 independent source · first recorded Sep 21, 2026. What these labels mean.

What the Qwen 3.8 Flash numbers mean

The fastest verified result on our board is 86.8 output tok/s on an DGX Spark at EXL3 3.05bpw with a 2048-token context. That is about 11.5 ms per generated token — 500 tokens takes roughly 5.8 seconds.

How much VRAM does Qwen 3.8 Flash need?

Qwen 3.8 Flash has 125.0B 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 Qwen 3.8 Flash's parameter count. Full benchmark board.

All benchmarks Methodology