Hardware · Orange Pi

What AI models can the Orange Pi AIpro (20T) (24 GB) run?

RAM: 24.0 GB
Usable for models: 21.1 GB
Memory bandwidth: 34 GB/s
Runs models on: CPU cores
Released: 2024

459 / 622 LLMs run Best: Gemma 4 26B-A4B 58 / 61 Image models run Best: FLUX 3 Image Klein 24 / 43 Video models run Best: Bernini-R A14B (video generation + editing) 66 / 70 Decision models run Best: Qwen3 Reranker 4B

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 21.1 GB of the 24.0 GB (what Linux leaves free for a model). Ascend DaVinci V300 AI core: 20 TOPS INT8 / 10 TFLOPS FP16, programmed with Huawei’s CANN tools. llama.cpp’s CANN backend doesn’t support the 310B yet, so llama.cpp and Ollama run on the four CPU cores.

More about this hardware

Maker	Orange Pi
Type	Single-board computer or edge AI board
Share of memory for models	88% (what Linux typically leaves free on a board this size)
Chip	Huawei Ascend 310B
AI accelerator	Ascend DaVinci V300 AI core: 20 TOPS INT8 / 10 TFLOPS FP16, programmed with Huawei’s CANN tools. llama.cpp’s CANN backend doesn’t support the 310B yet, so llama.cpp and Ollama run on the four CPU cores.
Boards with this chip	Orange Pi AIpro (20T) 12 GB / 24 GB
Memory type	LPDDR4X-4266 (Orange Pi doesn’t publish the bus width)
Bandwidth figure	34.1 GB/s, estimated, see the note below
Note	4 TaiShan V200M cores at 1.6 GHz. Bandwidth assumes a 64-bit bus; speeds assume llama.cpp on the CPU at about 30% of it (estimate). Its NPU only helps with models converted for Huawei’s tools.
Specifications	www.orangepi.org

Apps that run models well on the Orange Pi AIpro (20T) (24 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 Orange Pi AIpro (20T) (24 GB)

Orange Pi AIpro (20T)Orange Pi · Board · 12 or 24 GBProduct page (opens in a new tab)

Best LLMs for the Orange Pi AIpro (20T) (24 GB)

Quality	Model	Quant	Memory	Speed	Runs
89Excellent quality (A+) · 89 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking.	Gemma 4 26B-A4BGoogle	`Q5_K_M`	21.0 GB	3.7 tok/s	Slowtight fit
90Excellent quality (A+) · 90 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.8 27BAlibaba	`Q4_K_M`	19.8 GB	0.6 tok/s	Very slowtight fit
83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	Raon-1 mini Preview (30B-A4B)KRAFTON	`Q4_K_M`	19.9 GB	4.0 tok/s	Slowtight fit
89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.6 27BAlibaba	`Q4_K_M`	19.8 GB	0.6 tok/s	Very slowtight fit
78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.6 35B-A3BAlibaba	`Q3_K_M`	20.1 GB	7.3 tok/s	Slowtight fit
78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	MiniCPM-V 4.7 35B-A3BOpenBMB	`Q3_K_M`	19.7 GB	7.3 tok/s	Slowtight fit
88Excellent quality (A+) · 88 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.5 27BAlibaba	`Q4_K_M`	19.8 GB	0.6 tok/s	Very slowtight fit
78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.5 35B-A3BAlibaba	`Q3_K_M`	20.1 GB	7.3 tok/s	Slowtight fit
77Great quality (A) · 77 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	Nex N2 Mini (35B-A3B)Nex-AGI	`Q3_K_M`	19.7 GB	7.3 tok/s	Slowtight fit
87Excellent quality (A+) · 87 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	Muse Glimmer 30BMeta	`Q4_K_M`	20.6 GB	0.6 tok/s	Very slowtight fit
76Great quality (A) · 76 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	KAT-Coder V2.5 Dev (35B-A3B)Kuaishou	`Q3_K_M`	19.4 GB	7.3 tok/s	Slowtight fit
76Great quality (A) · 76 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	Sage-A1.5 35B-A3BInternScience	`Q3_K_M`	20.1 GB	7.3 tok/s	Slowtight fit

Best Image models for the Orange Pi AIpro (20T) (24 GB)

Quality	Model	Quant	Memory	Speed	Runs
90Excellent quality (A+) · 90 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	FLUX 3 Image KleinBlack Forest Labs	`Q8 GGUF`	12.8 GB	26.9 min	Very slow
90Excellent quality (A+) · 90 of 100: how capable the model is at FP8, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	Krea 2 TurboKrea	`FP8`	20.2 GB	15.9 min	Very slowtight fit
84Great quality (A) · 84 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	ERNIE-Image TurboBaidu	`Q8 GGUF`	15.0 GB	15.1 min	Very slow
83Great quality (A) · 83 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	Z-Image TurboAlibaba (Tongyi)	`Q8 GGUF`	13.0 GB	11.8 min	Very slow
87Excellent quality (A+) · 87 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	ERNIE-ImageBaidu	`Q8 GGUF`	18.9 GB	1.7 h	Very slowtight fit
82Great quality (A) · 82 of 100: how capable the model is at FP8, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	FLUX.2 [klein] 9BBlack Forest Labs	`FP8`	19.9 GB	6.7 min	Very slowtight fit

Best Video models for the Orange Pi AIpro (20T) (24 GB)

Quality	Model	Quant	Memory	Speed	Runs
83Great quality (A) · 83 of 100: how capable the model is at FP8, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking.	Bernini-R A14B (video generation + editing)ByteDance	`FP8`	20.3 GB	46.6 h	Very slowtight fit · encoder unloaded after use
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.	HY-OmniWeaving (8.3B, unified video)Tencent	`FP8`	15.7 GB	34.3 h	Very slowencoder unloaded after use
70Great quality (A) · 70 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking.	Wan 2.2 S2V 14B (speech to video)Alibaba	`Q4 GGUF`	15.6 GB	86.3 h	Very slowencoder unloaded after use
68Good quality (B) · 68 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	Wan 2.1 I2V 14B 480PAlibaba	`Q4 GGUF`	14.2 GB	77.7 h	Very slowencoder unloaded after use
68Good quality (B) · 68 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	Kandinsky 6.0 Video Lite (3B, video + audio)Sber AI	`FP16`	14.7 GB	22.3 h	Very slowencoder unloaded after use

Best Decision models for the Orange Pi AIpro (20T) (24 GB)

Quality	Model	Quant	Memory	Speed	Runs
90Excellent quality (A+) · 90 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3 Reranker 4BAlibaba	`Q8_0`	6.9 GB	69.1 s	Very slow
90Excellent quality (A+) · 90 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.	Qwen3 Reranker 8BAlibaba	`Q8_0`	11.3 GB	140.7 s	Very slow
90Excellent quality (A+) · 90 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.	Skywork Reward V2 Llama 3.1 8BSkywork	`Q8_0`	10.4 GB	278.3 s	Very slow
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`	3.2 GB	10.3 s	Very slow
88Excellent quality (A+) · 88 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.	zerank-2ZeroEntropy	`Q8_0`	6.9 GB	69.1 s	Very slow
78Great quality (A) · 78 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.	OpenAI Privacy FilterOpenAI	`Q8_0`	2.4 GB	468 ms	Slow
86Excellent quality (A+) · 86 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.	Shieldstral 1.0 3BMistral AI	`Q8_0`	6.6 GB	66.1 s	Very slow
84Great quality (A) · 84 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.	Jina Reranker v3Jina AI	`Q8_0`	3.2 GB	10.3 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.