M5Stack Module LLM (AX630C)

A tiny 54mm module with an Axera AX630C NPU, running StackFlow.

Fastest measured run

12.52tok/s

Qwen2.5 0.5B Instruct · W4A16 · StackFlow · source ↗

For the M5Stack LLM Module Kit, listed at $79.9. All options · What else you need

Where to buy.

M5Stack LLM Module Kit (AX630C)$79.9
regular price; out of stock when checked · price source ↗ (2026-10-17)
Check price on Amazon
M5Stack Module LLM (as tested in paper)$99.9
price quoted in arXiv paper · price source ↗ (2026-04-24)
Check price on Amazon

The kit has its own storage, mic and speaker. Compare complete starter kits

Good to know: The LLM Module Kit has 32GB of eMMC and a microphone and speaker on board, so no storage card is needed. source ↗

M5Stack CoreS3 (host controller)
Optional: host controller
M5Stack designs the LLM Module Kit to pair with an M5 host controller; CoreS3 is one of them. source ↗
Check price on Amazon

Board + required parts, approximate

≈ $79.90

M5Stack LLM Module Kit $79.90. Sourced prices at time of writing; they change often, shipping and tax not included.

Model	Size	Quant	Runtime	Gen tok/s	Prompt tok/s	TTFT ms	Source
Qwen2.5 0.5B Instruct	500M	`W4A16`	StackFlow	12.52	–	442.95	source ↗ accessed 2026-10-17
Qwen2.5 0.5B Instruct	500M	`W8A16`	StackFlow	10.32	–	359.8	source ↗ accessed 2026-10-17
Qwen3 0.6B	600M	`W8A16`	StackFlow	10.28	–	361.81	source ↗ accessed 2026-10-17
InternVL3 1B VLM	1B	`W8A16`	StackFlow	9.78	–	534.95	source ↗ accessed 2026-10-17
Qwen2.5 1.5B Instruct	1.5B	`W4A16`	StackFlow	4.63	–	1219.54	source ↗ accessed 2026-10-17
Llama 3.2 1B Instruct	1B	`W8A16`	StackFlow	4.48	–	891	source ↗ accessed 2026-10-17
DeepSeek-R1-Distill-Qwen 1.5B	1.5B	`W8A16`	StackFlow	3.57	–	1075.04	source ↗ accessed 2026-10-17
Qwen2.5 1.5B Instruct	1.5B	`W8A16`	StackFlow	3.57	–	3056.54	source ↗ accessed 2026-10-17
Llama 3.2 1B	1B	`not reported`	StackFlow	3.44	–	–	source ↗ 2026-04-24
Qwen2.5 1.5B Instruct	1.5B	`not reported`	StackFlow	2.55	–	–	source ↗ 2026-04-24
DeepSeek-R1-Distill-Qwen 1.5B	1.5B	`not reported`	StackFlow	2.42	–	–	source ↗ 2026-04-24