Hardware · NVIDIA
What AI models can the NVIDIA Jetson Orin Nano 4GB (4 GB) run?
- Unified memory
- 4.0 GB
- Usable for models
- 2.4 GB
- Memory bandwidth
- 51 GB/s
- Runs models on
- GPU (CUDA)
- Released
- 2023
Models run on the GPU (CUDA) here, sharing the board’s memory with the system. We plan around 2.4 GB of the 4.0 GB (what Linux leaves free for a model). Ampere GPU with 512 CUDA cores and 16 Tensor Cores: 34 TOPS (INT8, sparse) in Super mode.
More about this hardware
| Maker | NVIDIA |
|---|---|
| Type | Single-board computer or edge AI board |
| GPU share of memory | 60% (what the GPU can typically be given for a model) |
| Chip | NVIDIA Jetson Orin Nano 4GB |
| AI accelerator | Ampere GPU with 512 CUDA cores and 16 Tensor Cores: 34 TOPS (INT8, sparse) in Super mode. |
| Boards with this chip | Jetson Orin Nano 4GB module on a carrier board |
| Memory type | 4 GB LPDDR5, 64-bit (51 GB/s in Super mode, 34 GB/s before JetPack 6.2) |
| Bandwidth figure | 51.0 GB/s, as the maker states it |
| Note | Sold as a module (NVIDIA’s developer kit has the 8 GB version). Figures are for Super mode, which needs JetPack 6.2 flashed with a Super configuration. About 2.5 GB is free for models once Ubuntu is running (estimate). GPU figure scaled from the 8 GB module (half the CUDA cores). |
| Specifications | www.nvidia.com |
Apps that run models well on the NVIDIA Jetson Orin Nano 4GB (4 GB)
Linux on ARM boards
Best LLMs for the NVIDIA Jetson Orin Nano 4GB (4 GB)
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 53Fair quality (C) · 53 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | LFM3 1.7B | Q4_K_M | 1.9 GB | 36.2 tok/s | Runs great |
| 50Fair quality (C) · 50 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | MiniCPM5 2B | Q3_K_M | 2.4 GB | 32.1 tok/s | Runs welltight fit |
| 47Fair quality (C) · 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | LFM3-VL 2B | Q4_K_M | 2.1 GB | 32.6 tok/s | Runs welltight fit |
| 41Fair quality (C) · 41 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. | PaddleOCR-VL 1.6 0.9B | Q6_K | 1.7 GB | 45.2 tok/s | Runs great |
| 40Fair quality (C) · 40 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | MiniCPM-OCR 1.0 1.3B | Q5_K_M | 1.8 GB | 39.8 tok/s | Runs great |
| 38Basic quality (D) · 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | MiniCPM-V 4.7 1B | Q4_K_M | 1.8 GB | 40.4 tok/s | Runs great |
| 35Basic quality (D) · 35 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | MiniCPM5 1B | Q6_K | 1.9 GB | 40.2 tok/s | Runs great |
| 38Basic quality (D) · 38 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. | TMax 2B | Q4_K_M | 2.0 GB | 33.3 tok/s | Runs great |
| 33Basic quality (D) · 33 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | Qwen3.5 2B | Q3_K_M | 2.0 GB | 35.7 tok/s | Runs great |
| 31Basic quality (D) · 31 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | LFM2.5 1.2B Instruct | Q5_K_M | 1.7 GB | 44.2 tok/s | Runs great |
| 36Basic quality (D) · 36 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | LFM2.5 2.6B | Q3_K_M | 2.2 GB | 30.0 tok/s | Runs welltight fit |
| 36Basic quality (D) · 36 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | Jamba Reasoning 3B | Q3_K_M | 2.4 GB | 25.3 tok/s | Runs welltight fit |
Best Image models for the NVIDIA Jetson Orin Nano 4GB (4 GB)
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 39Basic quality (D) · 39 of 100: how capable the model is at FP8, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | SDXL Turbo | FP8 | 1.6 GB | 4.6 s | Runs greatCPU offload |
| 50Fair quality (C) · 50 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. | PixArt-Σ XL 1024 | FP16 | 2.3 GB | 52 s | Slowtight fit · CPU offload |
| 30Basic quality (D) · 30 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. | LCM DreamShaper v7 | FP16 | 1.3 GB | 8.6 s | Runs greatCPU offload |
| 28Basic quality (D) · 28 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. | SD Turbo | FP16 | 1.3 GB | 2.7 s | Runs greatCPU offload |
| 42Fair quality (C) · 42 of 100: how capable the model is at FP16, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking. | Realistic Vision V5.1 | FP16 | 1.3 GB | 41 s | SlowCPU offload |
| 40Fair quality (C) · 40 of 100: how capable the model is at FP16, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking. | DreamShaper 8 | FP16 | 1.3 GB | 41 s | SlowCPU offload |
Best Video models for the NVIDIA Jetson Orin Nano 4GB (4 GB)
None of the video models in our catalog fit in this much memory.
Best Decision models for the NVIDIA Jetson Orin Nano 4GB (4 GB)
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. | GLiNER2.5 Decide | Q8_0 | 1.2 GB | 184 ms | Runs well |
| 86Excellent quality (A+) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | Jina Reranker v3.5 | Q8_0 | 1.7 GB | 429 ms | Slow |
| 68Good quality (B) · 68 of 100: how capable the model is at F16, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | Llama Prompt Guard 2 22M | F16 | 0.8 GB | 64 ms | Runs great |
| 70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | GTE Reranker ModernBERT Base | Q8_0 | 0.8 GB | 114 ms | Runs great |
| 74Great quality (A) · 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | GLiClass ModernBERT Large v3 | Q8_0 | 1.1 GB | 151 ms | Runs well |
| 84Great quality (A) · 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. | Jina Reranker v3 | Q8_0 | 1.7 GB | 429 ms | Slow |
| 66Good quality (B) · 66 of 100: how capable the model is at F16, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | GLiClass Base v3 | F16 | 1.0 GB | 84 ms | Runs great |
| 68Good quality (B) · 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. | Granite Embedding Reranker English R2 | Q8_0 | 0.8 GB | 114 ms | Runs great |
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.