Qwen3.5-0.8B

Qwen family · 0.8B params

5recorded runs

554.5best tok/s

5verified

Evidence: 5 sourced runs · 1 independent source · first recorded Aug 10, 2026 · last recorded Aug 10, 2026. What these labels mean.

What the Qwen3.5-0.8B numbers mean

The fastest verified result on our board is 554.5 output tok/s on an Radeon AI Pro R9700 at MQ4 with a 128-token context. That is about 1.8 ms per generated token — 500 tokens takes roughly 0.9 seconds. Across the 5 records we hold for this model the range is 273.6–554.5 tok/s on 5 GPUs, because quantisation, context length and the engine build all move the number.

How much VRAM does Qwen3.5-0.8B need?

Qwen3.5-0.8B has 0.8B 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.

Quantisation	Estimated VRAM	Smallest card that fits
Q4_K_M (4-bit)	1.3 GB	8 GB
Q8_0 (8-bit)	1.8 GB	8 GB
FP16 (16-bit)	2.6 GB	8 GB

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.

Fastest GPUs for this model (best tok/s)

Models with a similar footprint

Model	Params	Q4_K_M weights
Qwen3.5-0.8B-MTP	800M	0.5 GB
gemma-3-1b-it	1B	0.6 GB
LFM2.5-1.2B-Instruct	1B	0.6 GB
Llama-3.2-1B-Instruct	1B	0.6 GB
MiniCPM5-1B	1B	0.6 GB
Qwen3-0.6B	1B	0.6 GB

Ranked by distance from Qwen3.5-0.8B's parameter count. Full benchmark board.

All benchmarks Methodology

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
Radeon AI Pro R9700 · 32 GB	Qwen3.5-0.8B	MQ4	hipfire 0.3.0+44eb828cc804	128	554.5 ✓	—	—	src
RX 7900 XTX · 24 GB	Qwen3.5-0.8B	MQ4	hipfire 0.3.0+44eb828cc804	128	550.2 ✓	—	—	src
RX 9070 XT · 16 GB	Qwen3.5-0.8B	MQ4	hipfire 0.3.0+44eb828cc804	128	545.3 ✓	—	—	src
RX 6900 XT · 16 GB	Qwen3.5-0.8B	MQ4	hipfire 0.3.0+32d27a87d7f0	128	346.1 ✓	—	—	src
RX 5700 XT · 8 GB	Qwen3.5-0.8B	MQ4	hipfire 0.3.0+32d27a87d7f0	128	273.6 ✓	—	—	src