What AI models can the Arduino Portenta X8 (2 GB) run?

RAM: 2.0 GB
Usable for models: 1.2 GB
Memory bandwidth: 12 GB/s
Runs models on: CPU cores
Released: 2022

36 / 622 LLMs run Best: Falcon-H1 Tiny 90M Instruct 0 / 61 Image models run None fit 0 / 43 Video models run None fit 25 / 70 Decision models run Best: Llama Prompt Guard 2 86M

Models run on the CPU cores here, so the memory’s speed and how fast those cores can read it set the pace: small models (about 1-4B) are usable, bigger ones are slow. We plan around 1.2 GB of the 2.0 GB (what Linux leaves free for a model). No NPU. An STM32H747 microcontroller handles real-time I/O next to the Linux processor.

More about this hardware

Maker	Arduino
Type	Single-board computer or edge AI board
Share of memory for models	60% (what Linux typically leaves free on a board this size)
Chip	NXP i.MX 8M Mini
AI accelerator	No NPU. An STM32H747 microcontroller handles real-time I/O next to the Linux processor.
Boards with this chip	Arduino Portenta X8
Memory type	2 GB LPDDR4
Bandwidth figure	12.0 GB/s, estimated, see the note below
Note	4× Cortex-A53 at 1.8 GHz running Linux with Docker containers. Only the smallest models (up to about 0.5B) are practical (estimate, assuming a 32-bit bus at 3000 MT/s).
Specifications	www.arduino.cc

Apps that run models well on the Arduino Portenta X8 (2 GB)

Linux on ARM boards

For a typical model that fits in memory. Each model page picks the best app for that model on this hardware. All apps

1 product with the Arduino Portenta X8 (2 GB)

Portenta X8Arduino · Board · 2 GBProduct page (opens in a new tab)

Best LLMs for the Arduino Portenta X8 (2 GB)

Quality	Model	Quant	Memory	Speed	Runs
2Basic quality (D) · 2 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking.	Falcon-H1 Tiny 90M InstructTII	`Q2_K`	0.8 GB	44.8 tok/s	Runs great
2Basic quality (D) · 2 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking.	SmolLM2 135M InstructHugging Face	`Q2_K`	0.9 GB	31.0 tok/s	Runs great
1Basic quality (D) · 1 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking.	SmolLM 135M InstructHugging Face	`Q2_K`	0.8 GB	31.0 tok/s	Runs great
7Basic quality (D) · 7 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking.	LFM2.5 230MLiquid AI	`Q2_K`	0.9 GB	17.5 tok/s	Runs well
9Basic quality (D) · 9 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking.	LFM2.5 350MLiquid AI	`Q2_K`	0.9 GB	11.5 tok/s	Runs well
18Basic quality (D) · 18 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.5 0.8BAlibaba	`Q2_K`	1.1 GB	4.6 tok/s	Slowtight fit
8Basic quality (D) · 8 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking.	Granite 4.0 H 350MIBM	`Q2_K`	0.9 GB	11.9 tok/s	Runs well
8Basic quality (D) · 8 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking.	LFM2 350MLiquid AI	`Q2_K`	0.9 GB	11.5 tok/s	Runs well
4Basic quality (D) · 4 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking.	SmolVLM 256M InstructHugging Face	`Q2_K`	1.0 GB	15.5 tok/s	Runs well
4Basic quality (D) · 4 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking.	Gemma 3 270MGoogle	`Q2_K`	0.8 GB	14.9 tok/s	Runs well
8Basic quality (D) · 8 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking.	LFM2-VL 450MLiquid AI	`Q2_K`	1.0 GB	9.0 tok/s	Slow
15Basic quality (D) · 15 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking.	LFM2 700MLiquid AI	`Q3_K_M`	1.2 GB	4.3 tok/s	Slowtight fit

Best Image models for the Arduino Portenta X8 (2 GB)

None of the image models in our catalog fit in this much memory.

Best Video models for the Arduino Portenta X8 (2 GB)

None of the video models in our catalog fit in this much memory.

Best Decision models for the Arduino Portenta X8 (2 GB)

Quality	Model	Quant	Memory	Speed	Runs
76Great quality (A) · 76 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking.	Llama Prompt Guard 2 86MMeta	`Q8_0`	1.0 GB	21.6 s	Very slow
75Great quality (A) · 75 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.	mxbai-rerank-base v2Mixedbread	`Q6_K`	1.1 GB	35.1 s	Very slowtight fit
74Great quality (A) · 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.	GLiClass ModernBERT Large v3Knowledgator	`Q8_0`	1.1 GB	15.5 s	Very slowtight fit
70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	GTE Reranker ModernBERT BaseAlibaba	`Q8_0`	0.8 GB	11.6 s	Very slow
68Good quality (B) · 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	Llama Prompt Guard 2 22MMeta	`Q8_0`	0.7 GB	5.5 s	Very slow
72Great quality (A) · 72 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	DeBERTa v3 Large Zero-shot v2.0Moritz Laurer	`Q8_0`	1.2 GB	17.0 s	Very slowtight fit
70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	Jina Reranker v2 Base MultilingualJina AI	`Q8_0`	1.0 GB	21.6 s	Very slow
68Good quality (B) · 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking.	Granite Embedding Reranker English R2IBM	`Q8_0`	0.8 GB	11.6 s	Very slow

Best LLMs by task on this hardware Open in Model Finder

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.