AI station / mini-PC
Beelink SER9 (Ryzen AI 9, 32 GB) — Local LLM Performance
Beelink SER9 (Ryzen AI 9, 32 GB) is a Mini PC running Windows or Linux, with 32 GB unified memory.
Specifications
- Memory for models
- 32 GB unified
- Memory bandwidth
- 120 GB/s
- Memory upgradeable
- Yes
- Architecture
- Zen 5 + RDNA 3.5 "Strix Point"
- Form factor
- Mini PC
- Operating system
- Windows or Linux
- Released
- 2025
- Llama 3.1 8B Q4_K_M
- 9.0–19 tok/s
- Price
- $859 (checked 2026-07-06)
Worth knowing before you buy
- Shares system memory with the iGPU, so the usable model budget is well below the nominal 32 GB.
- Memory bandwidth, not capacity, is the limit here — expect single-digit tokens/sec on large models.
Models on the Beelink SER9 32 GB
- Yi 1.5 FamilyYes — comfortably
- VibeThinkerYes — comfortably
- Aya 3B (Tiny Aya)Yes — comfortably
- StarCoder 2Yes — comfortably
- SmolLM3Yes — comfortably
- SmolLM2Yes — comfortably
- Qwen3-CoderYes
- Qwen3.8Yes — comfortably
- Qwen 3.6Yes, but it is tight
- Qwen 3.5Yes, but it is tight
- Qwen 3Yes — comfortably
- Qwen 2.5 VLYes — comfortably
- Qwen 2.5 FamilyYes — comfortably
- Poolside Laguna XS 2.1Yes, but it is tight
- Phi-4 MiniYes — comfortably
- Phi-4 FamilyYes — comfortably
- Phi 3.5 FamilyYes — comfortably
- OLMo 2Yes — comfortably
- North Mini CodeYes
- Nex-N2Yes, but it is tight
- Nex-N2.5Yes, but it is tight
- Nemotron Cascade 2Yes
- Nemotron 70BTechnically yes, but not recommended
- Nemotron 3 Nano OmniYes
- Mistral Small 3.1Yes — comfortably
- Mistral Small 3.2Yes — comfortably
- Mistral FamilyYes — comfortably
- MinistralYes — comfortably
- Ministral 3Yes — comfortably
- MiniCPM-VYes — comfortably
- Magistral SmallYes — comfortably
- Llama 3.3Technically yes, but not recommended
- Llama 3.2 VisionYes — comfortably
- Llama 3.2 FamilyYes — comfortably
- Llama 3.1 FamilyYes — comfortably
- LFM2.5Yes
- InternLM 3Yes — comfortably
- IBM Granite 4.0Yes, but it is tight
- IBM Granite 4.2Yes — comfortably
- IBM Granite 4.1Yes — comfortably
- Granite 3.0Yes — comfortably
- GPT-OSSYes — comfortably
- GLM-4.7 / GLM-Z1Yes — comfortably
- GLM-5 / GLM-5.1No
- Gemma 4Yes — comfortably
- Gemma 3nYes — comfortably
- Gemma 3Yes — comfortably
- Gemma 2 FamilyYes — comfortably
- Falcon 3Yes — comfortably
- EXAONE 3.5Yes — comfortably
- DevstralYes — comfortably
- DeepSeek R1Yes — comfortably
- DeepSeek-OCRYes — comfortably
- Cosmos 3Yes — comfortably
- Command R FamilyYes — comfortably
- Cogito v1Yes — comfortably
- CodestralYes — comfortably
- Bonsai 27BYes — comfortably
- BitNet b1.58Yes — comfortably
- Aya ExpanseYes — comfortably