Asustor AS6508T

Hardware · Intel

What AI models can the Intel Atom C3538 / C3558 NAS (64 GB) run?

RAM
64.0 GB
Usable for models
54.4 GB
Memory bandwidth
19 GB/s
Runs models on
CPU cores
Released
2017
520 / 622 LLMs run Best: Nanbeige4.3 33B 61 / 61 Image models run Best: Krea 2 Turbo 42 / 43 Video models run Best: LTX-2.3 69 / 70 Decision models run Best: Qwen3 Reranker 4B

A NAS with this chip runs models on its processor cores, so memory speed sets the pace: small models (about 1-4B) answer at a usable speed, bigger ones are slow. We plan around 54.4 GB for models, what is left after the NAS system and its storage services. Models run in a container (Docker) or the maker’s app store, and speeds assume the NAS isn’t busy with storage work.

More about this hardware

MakerIntel
TypeNAS (network attached storage) processor
Share of memory for models85% (what the NAS system and its storage services typically leave free at this size)
ChipIntel Atom C3538 / C3558
AI acceleratorNone. Models run on the processor cores.
NAS models with this chipAsustor AS6508T · Asustor AS6510T · Asustor AS6504RS · Asustor AS6504RD · Asustor AS6512RD · Synology RS820RP+ · Synology RS820+ · Synology RS2818RP+ · Synology RS2418RP+ · Synology RS2418+ · Synology DS2419+II · Synology DS2419+ · Synology DS1819+ · Synology DS1618+ · Synology DVA3221 · Synology DVA3219 · TerraMaster T9-450 · TerraMaster T12-450 · TerraMaster U8-450 · Buffalo TeraStation WS5020 IoT 2025 desktop (WS5220DN / WS5420DN) · Buffalo TeraStation WS5020 IoT 2025 rackmount (WS5420RN) · Netgear ReadyNAS 426 / 428 (RN426, RN428) · iXsystems TrueNAS Mini X
Memory typeSystem RAM, shared with the NAS operating system and its storage services
Bandwidth figure19.2 GB/s, estimated, see the note below
NoteOne profile for every NAS built on this chip. Memory speed is an estimate from the chip’s supported memory in a typical NAS fitting; a unit with one memory module runs slower than one with two. Speed assumes the NAS is not busy with storage work.
Specificationswww.asustor.com

Apps that run models well on the Intel Atom C3538 / C3558 NAS (64 GB)

Linux

  1. OllamaOne command with Ollama’s own build of this model.
  2. llama.cppUsually the first to support new model architectures.
  3. KoboldCpp

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

5 products with the Intel Atom C3538 / C3558 NAS (64 GB)

Best LLMs for the Intel Atom C3538 / C3558 NAS (64 GB)

Quality Model Quant Memory Speed Runs
95Excellent quality (A+) · 95 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. Nanbeige4.3 33BBOSS Zhipin (Nanbeige) Q8_0 41.8 GB 0.2 tok/s Very slow
95Excellent quality (A+) · 95 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.8 27BAlibaba Q8_0 34.9 GB 0.2 tok/s Very slow
94Excellent quality (A+) · 94 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 27BAlibaba Q8_0 34.9 GB 0.2 tok/s Very slow
87Excellent quality (A+) · 87 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. Flint (40B-A2.5B)Recursive Q5_K_M 33.4 GB 3.2 tok/s Slow
93Excellent quality (A+) · 93 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. Qwen3.5 27BAlibaba Q8_0 34.9 GB 0.2 tok/s Very slow
92Excellent quality (A+) · 92 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. Muse Glimmer 30BMeta Q8_0 36.6 GB 0.2 tok/s Very slow
91Excellent quality (A+) · 91 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. Gemma 4 31BGoogle Q8_0 39.7 GB 0.2 tok/s Very slow
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. TMax 27BAi2 Q8_0 34.5 GB 0.2 tok/s Very slow
87Excellent quality (A+) · 87 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. Nemotron 3.5 Bolt 74B-A9BNVIDIA Q4_K_M 50.7 GB 1.1 tok/s Very 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. Qwen3.6 35B-A3BAlibaba Q4_K_M 27.0 GB 3.2 tok/s Slow
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 35B-A3BAlibaba Q4_K_M 27.0 GB 3.2 tok/s Slow
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. MiniCPM-V 4.7 35B-A3BOpenBMB Q4_K_M 26.6 GB 3.2 tok/s Slow

Best Image models for the Intel Atom C3538 / C3558 NAS (64 GB)

Quality Model Quant Memory Speed Runs
92Excellent quality (A+) · 92 of 100: how capable the model is at FP16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. Krea 2 TurboKrea FP16 37.4 GB 29.2 min Very slow
92Excellent quality (A+) · 92 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. Qwen-Image 2.1Alibaba Q8 GGUF 41.6 GB 7.8 h Very slow
89Excellent quality (A+) · 89 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. Qwen-Image 2512Alibaba Q8 GGUF 36.5 GB 11.7 h Very slow
84Great quality (A) · 84 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. FLUX.2 [klein] 9BBlack Forest Labs FP16 37.1 GB 12.2 min Very slow
88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. ERNIE-ImageBaidu FP16 34.0 GB 2.5 h Very slow
93Excellent quality (A+) · 93 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. Fibo 2.0 DevBria AI FP16 27.7 GB 2.5 h Very slow

Best Video models for the Intel Atom C3538 / C3558 NAS (64 GB)

Quality Model Quant Memory Speed Runs
85Excellent quality (A+) · 85 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. LTX-2.3Lightricks Q8 GGUF 45.6 GB 52.0 h Very slow
85Excellent quality (A+) · 85 of 100: how capable the model is at FP16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. Bernini-R A14B (video generation + editing)ByteDance FP16 46.0 GB 81.4 h Very slow
85Excellent quality (A+) · 85 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. Wan 2.2 I2V A14BAlibaba Q8 GGUF 39.3 GB 207.9 h Very slow
85Excellent quality (A+) · 85 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. Wan 2.2 T2V A14BAlibaba Q8 GGUF 39.3 GB 207.9 h Very slow
87Excellent quality (A+) · 87 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. LTX-2.5 (22B, video + audio)Lightricks Q8 GGUF 49.8 GB 11.8 h Very slowtight fit

Best Decision models for the Intel Atom C3538 / C3558 NAS (64 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 80/100 (quality, fit, speed and memory use) sets the ranking. Qwen3 Reranker 4BAlibaba Q8_0 9.5 GB 115.2 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 13.9 GB 234.6 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 463.8 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 9.5 GB 115.2 s Very 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. Jina Reranker v3.5Jina AI Q8_0 5.8 GB 17.2 s Very 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 9.3 GB 110.3 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 v3Jina AI Q8_0 5.8 GB 17.2 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. GLiDE 27BFastino Q8_0 34.8 GB 403.3 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.

Was this page helpful?