Hardware · Rockchip

What AI models can the Rockchip RK3576 (4 GB) run?

RAM: 4.0 GB
Usable for models: 2.8 GB
Memory bandwidth: 17 GB/s
Runs models on: CPU cores
Released: 2024

126 / 622 LLMs run Best: MiniCPM5 2B 7 / 61 Image models run Best: SD Turbo 1 / 43 Video models run Best: AnimateDiff v3 (SD 1.5) 41 / 70 Decision models run Best: OpenAI Privacy Filter

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 2.8 GB of the 4.0 GB (what Linux leaves free for a model). 6 TOPS NPU (INT8), programmed with Rockchip’s RKNN and RKLLM toolkits. llama.cpp and Ollama use the CPU.

More about this hardware

Maker	Rockchip
Type	Single-board computer or edge AI board
Share of memory for models	70% (what Linux typically leaves free on a board this size)
Chip	Rockchip RK3576
AI accelerator	6 TOPS NPU (INT8), programmed with Rockchip’s RKNN and RKLLM toolkits. llama.cpp and Ollama use the CPU.
Boards with this chip	Seeed reComputer RK3576 · Radxa ROCK 4D · FriendlyElec NanoPi M5 · Luckfox Omni3576 · KICKPI K7
Memory type	LPDDR4X-4266 or LPDDR5-5500, 32-bit (17-22 GB/s)
Bandwidth figure	17.1 GB/s, the chip’s rated memory speed
Note	4× Cortex-A72 at 2.2 GHz plus 4× Cortex-A53, Mali-G52 GPU, up to 16 GB. Half the memory bus of the RK3588 (32-bit), so models generate at roughly half its speed. Speeds assume llama.cpp on the CPU at about 45% of the rated memory speed (estimate).
Specifications	www.rock-chips.com

Apps that run models well on the Rockchip RK3576 (4 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

Best LLMs for the Rockchip RK3576 (4 GB)

Quality	Model	Quant	Memory	Speed	Runs
57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking.	MiniCPM5 2BOpenBMB	`Q4_K_M`	2.7 GB	5.2 tok/s	Slowtight fit
30Basic quality (D) · 30 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	Unlimited OCR 2 (3B-A0.6B)Baidu	`Q2_K`	2.7 GB	32.3 tok/s	Runs welltight fit
46Fair quality (C) · 46 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking.	Spark-X2.6 4BSparkLLM	`Q2_K`	2.8 GB	5.2 tok/s	Slowtight fit
28Basic quality (D) · 28 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking.	Unlimited OCR (3B-A0.6B)Baidu	`Q2_K`	2.7 GB	32.3 tok/s	Runs welltight fit
39Basic quality (D) · 39 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking.	MiniCPM-OCR 1.0 1.3BOpenBMB	`Q4_K_M`	1.7 GB	10.1 tok/s	Runs well
39Basic quality (D) · 39 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking.	PaddleOCR-VL 1.6 0.9BBaidu	`Q4_K_M`	1.6 GB	13.7 tok/s	Runs well
39Basic quality (D) · 39 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking.	LFM3 1.7BLiquid AI	`Q2_K`	1.6 GB	12.5 tok/s	Runs well
42Fair quality (C) · 42 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking.	TMax 4BAi2	`Q2_K`	2.8 GB	5.1 tok/s	Slowtight fit
41Fair quality (C) · 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking.	LFM2.5 2.6BLiquid AI	`Q4_K_M`	2.6 GB	4.9 tok/s	Slowtight fit
38Basic quality (D) · 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.5 2BAlibaba	`Q4_K_M`	2.3 GB	5.8 tok/s	Slow
35Basic quality (D) · 35 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking.	LFM3-VL 2BLiquid AI	`Q2_K`	1.7 GB	11.2 tok/s	Runs well
41Fair quality (C) · 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking.	Jamba Reasoning 3BAI21 Labs	`Q4_K_M`	2.8 GB	4.1 tok/s	Slowtight fit

Best Image models for the Rockchip RK3576 (4 GB)

Quality	Model	Quant	Memory	Speed	Runs
28Basic quality (D) · 28 of 100: how capable the model is at FP16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking.	SD TurboStability AI	`FP16`	2.6 GB	15 s	Runs welltight fit · encoder unloaded after use
42Fair quality (C) · 42 of 100: how capable the model is at FP16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking.	Realistic Vision V5.1SG161222	`FP16`	2.5 GB	3.8 min	Very slowtight fit · encoder unloaded after use
40Fair quality (C) · 40 of 100: how capable the model is at FP16, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking.	DreamShaper 8Lykon	`FP16`	2.5 GB	3.8 min	Very slowtight fit · encoder unloaded after use
30Basic quality (D) · 30 of 100: how capable the model is at FP16, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking.	LCM DreamShaper v7Tsinghua (LCM)	`FP16`	2.5 GB	48 s	Slowtight fit · encoder unloaded after use
50Fair quality (C) · 50 of 100: how capable the model is at FP16, whatever it runs on. Match score 52/100 (quality, fit, speed and memory use) sets the ranking.	PixArt-Σ XL 1024PixArt	`FP16`	2.3 GB	5.9 min	Very slowCPU offload
35Basic quality (D) · 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 51/100 (quality, fit, speed and memory use) sets the ranking.	Stable Diffusion 1.5Stability AI / Runway	`FP16`	2.5 GB	3.8 min	Very slowtight fit · encoder unloaded after use

Best Video models for the Rockchip RK3576 (4 GB)

Quality	Model	Quant	Memory	Speed	Runs
35Basic quality (D) · 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 43/100 (quality, fit, speed and memory use) sets the ranking.	AnimateDiff v3 (SD 1.5)Shanghai AI Lab	`FP16`	2.6 GB	78.8 min	Very slowtight fit · CPU offload

Best Decision models for the Rockchip RK3576 (4 GB)

Quality	Model	Quant	Memory	Speed	Runs
78Great quality (A) · 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking.	OpenAI Privacy FilterOpenAI	`Q8_0`	2.4 GB	210 ms	Runs welltight fit
86Excellent quality (A+) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	Jina Reranker v3.5Jina AI	`Q8_0`	1.7 GB	4.6 s	Very slow
84Great quality (A) · 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	Jina Reranker v3Jina AI	`Q8_0`	1.7 GB	4.6 s	Very slow
80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	GLiNER2.5 DecideFastino	`Q8_0`	1.2 GB	2.0 s	Very slow
80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	FRIDA-DecisionsSber AI	`Q8_0`	1.7 GB	3.4 s	Very slow
80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3 Reranker 0.6BAlibaba	`Q8_0`	1.7 GB	4.6 s	Very slow
81Great quality (A) · 81 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	mxbai-rerank-large v2Mixedbread	`Q6_K`	2.1 GB	11.9 s	Very slow
76Great quality (A) · 76 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	Llama Prompt Guard 2 86MMeta	`Q8_0`	1.0 GB	2.3 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.