Products · IceWhale · NAS
IceWhale ZimaBoard SE 1664
- Processor
- Intel Celeron J3455 / N3450 (Apollo Lake)
- Memory (RAM)
- 16 GB
- Launch price
- $199
- Released
- 2026
Specifications
| Maker | IceWhale |
|---|---|
| Type | NAS |
| Processor | Intel Celeron J3455 / N3450 (Apollo Lake) NAS (16 GB) |
| Memory (RAM) | 16.0 GB |
| Usable for AI models | 12.0 GB |
| Memory bandwidth | 15 GB/s |
| FP16 compute | 0.0 TFLOPS |
| Runs models on | CPU cores |
| Drive bays | 2x SATA 6Gb/s with power (external drives); 64GB eMMC |
| NAS system | ZimaOS Plus |
| Containers (Docker) | Yes |
| Virtual machines | Yes |
| Memory upgrade | DDR3L SODIMM (pre-installed) · 1 slot · up to 16.0 GB per the maker |
| Graphics card options | PCIe: 1x PCIe 2.0 x4 (exposed on the board edge). External GPU: No Thunderbolt. Maker's OCuLink GPU dock is listed for ZimaBlade (the same hardware platform), not yet by…. |
| Notes | Intel Atom x7-E3950 (Apollo Lake, up to 2.0 GHz). 2016-era Atom quad-core without AVX and one DDR3L slot (16GB fitted, 16GB max): only very small models at low speed through the ZimaOS app store Ollama or Docker. ZimaOS-AI needs an NVIDIA card or an i5-1235U, which this board lacks. |
| Product page | shop.zimaspace.com (opens in a new tab) |
Running AI on the IceWhale ZimaBoard SE 1664
It can, with limits: expect small models only, at modest speed. Models run on its CPU cores, and this configuration leaves about 12.0 GB for them after the NAS system and its storage services. The grades below assume the NAS isn’t busy with backups or media work at the same time.
Apps you can install to run models
| App | How it installs |
|---|---|
| Ollama | From the NAS app store |
| Open WebUI | From the NAS app store |
| LocalAI | From the NAS app store |
Best LLMs for the IceWhale ZimaBoard SE 1664
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 79Great quality (A) · 79 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. | K2 Horizon 7B | Q6_K | 10.1 GB | 0.4 tok/s | Very slow |
| 78Great quality (A) · 78 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. | Gemma 4 12B | Q4_K_M | 9.5 GB | 0.4 tok/s | Very slow |
| 77Great quality (A) · 77 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | MiMo V2.6 Distill Qwen 9B | Q6_K | 9.5 GB | 0.4 tok/s | Very slow |
| 77Great quality (A) · 77 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | SenseNova Loop 9B Preview | Q6_K | 10.0 GB | 0.4 tok/s | Very slow |
| 76Great quality (A) · 76 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | Ornith 1.5 9B | Q6_K | 9.7 GB | 0.4 tok/s | Very slow |
| 75Great quality (A) · 75 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | Qwen3.5 9B | Q6_K | 9.7 GB | 0.4 tok/s | Very slow |
| 74Great quality (A) · 74 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | Ornith 1.0 9B | Q6_K | 9.7 GB | 0.4 tok/s | Very slow |
| 73Great quality (A) · 73 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | TMax 9B | Q6_K | 9.2 GB | 0.4 tok/s | Very slow |
| 70Great quality (A) · 70 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. | Nemotron 3.5 Bolt E4B | Q8_0 | 10.2 GB | 0.6 tok/s | Very slow |
| 69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | UI-Mate 9B | Q6_K | 9.5 GB | 0.4 tok/s | Very slow |
Best Image models for the IceWhale ZimaBoard SE 1664
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 84Great quality (A) · 84 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. | FLUX 3 Image Klein | Q4 GGUF | 10.2 GB | 78.0 min | Very slow |
| 76Great quality (A) · 76 of 100: how capable the model is at FP8, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | FLUX.2 [klein] 4B | FP8 | 10.2 GB | 10.3 min | Very slow |
| 78Great quality (A) · 78 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | Z-Image Turbo | Q4 GGUF | 10.1 GB | 34.2 min | Very slow |
| 76Great quality (A) · 76 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | Z-Image | Q4 GGUF | 10.1 GB | 3.9 h | Very slow |
| 79Great quality (A) · 79 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | ERNIE-Image Turbo | Q4 GGUF | 11.3 GB | 43.8 min | Very slowtight fit |
| 87Excellent quality (A+) · 87 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | ERNIE-Image | Q8 GGUF | 11.2 GB | 5.0 h | Very slowtight fit · encoder unloaded after use |
Best Video models for the IceWhale ZimaBoard SE 1664
| 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 62/100 (quality, fit, speed and memory use) sets the ranking. | Bernini-R A14B (video generation + editing) | FP8 | 10.7 GB | 160.1 h | Very slowtight fit · CPU offload |
| 49Fair quality (C) · 49 of 100: how capable the model is at FP8, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking. | LTX-Video 2B | FP8 | 10.8 GB | 4.3 h | Very slowtight fit |
| 74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | HY-OmniWeaving (8.3B, unified video) | FP8 | 7.8 GB | 117.8 h | Very slowCPU offload |
| 55Good quality (B) · 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | Wan 2.1 T2V 1.3B | FP16 | 8.4 GB | 47.9 h | Very slowencoder unloaded after use |
| 55Good quality (B) · 55 of 100: how capable the model is at FP16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | Wan 2.1 VACE 1.3B | FP16 | 10.1 GB | 51.9 h | Very slowencoder unloaded after use |
Best Decision models for the IceWhale ZimaBoard SE 1664
| 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 80/100 (quality, fit, speed and memory use) sets the ranking. | Qwen3 Reranker 4B | Q8_0 | 6.1 GB | 344.6 s | Very slow |
| 89Excellent quality (A+) · 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | Qwen3 Reranker 8B | Q6_K | 8.6 GB | 701.2 s | Very slow |
| 88Excellent quality (A+) · 88 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. | zerank-2 | Q8_0 | 6.1 GB | 344.6 s | Very slow |
| 90Excellent quality (A+) · 90 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. | Skywork Reward V2 Llama 3.1 8B | Q8_0 | 10.4 GB | 1,391.3 s | Very slowtight fit |
| 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. | Jina Reranker v3.5 | Q8_0 | 2.4 GB | 51.4 s | Very slow |
| 86Excellent quality (A+) · 86 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. | Shieldstral 1.0 3B | Q8_0 | 5.9 GB | 330.0 s | Very slow |
| 84Great quality (A) · 84 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. | Jina Reranker v3 | Q8_0 | 2.4 GB | 51.4 s | Very slow |
| 84Great quality (A) · 84 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. | Intern-Decision 4B | Q8_0 | 6.4 GB | 195.5 s | Very slow |
Other products with the Intel Celeron J3455 / N3450 (Apollo Lake) NAS (16 GB)
Spec sources
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.