Beelink GTR9 Pro Ryzen AI Max+ 395 128GB
128 GB unified · 256 GB/s
In short
The Beelink GTR9 Pro Ryzen AI Max+ 395 128GB has 128 GB of unified memory, of which the GPU can use 96 GB by default, with 256 GB/s of memory bandwidth. At 8k context it runs 72 of the 113 open models we track fully in memory; the most capable is DeepSeek V4 Flash at UD-Q2_K_XL, at about 11 tok/s (estimated). We have no current price for the Beelink GTR9 Pro Ryzen AI Max+ 395 128GB in Germany.
Updated . Speeds are estimates; how they are worked out.
Prices
Every retailer we track in Germany, cheapest first. Buy links may earn us a commission.
No retailer in this market lists it right now.
What it runs
Every model in the catalogue, with the best quant for this hardware. Estimates, not measurements.
| Model | Quant | Fits | Memory | Max context | Decode @8k | Prefill | Good for |
|---|---|---|---|---|---|---|---|
| DeepSeek V4 Flash DeepSeek | UD-Q2_K_XL | fits | 91.7 GB | 55k99k with q8 cache | 11tok/sest. | 325 tok/s | – |
| Qwen 3.8 Flash Next Alibaba Qwen | UD-Q3_K_XL | fits | 84.9 GB | 256k | 3tok/sest. | 167 tok/s | – |
| MiniMax M2.7 MiniMax | UD-Q2_K_XL | fits | 72.9 GB | 101k186k with q8 cache | 15tok/sest. | 735 tok/s | – |
| Qwen 3.8 27B Alibaba Qwen | Q8_0 | fits | 28.5 GB | 256k | 5tok/sest. | 364 tok/s | – |
| Qwen 3.5 122B-A10B Alibaba Qwen | Q6_K | fits | 95.2 GB | 33k53k with q8 cache | 10tok/sest. | 962 tok/s | – |
| Qwen 3.6 27B Alibaba Qwen | Q8_0 | fits | 28.1 GB | 256k | 5tok/sest. | 364 tok/s | – |
| gpt-oss 120B OpenAI | MXFP4 | fits | 60.2 GB | 128k | 25tok/sest. | 1631 tok/s | chat |
| Qwen3 Coder Next Alibaba Qwen | Q8_0 | fits | 80.1 GB | 256k | 19tok/sest. | 2430 tok/s | chat |
| Nemotron 3 Super 120B-A12B NVIDIA | Q5_K_M | fits | 91.0 GB | 256k | 9tok/sest. | 847 tok/s | – |
| Gemma 4 31B | Q8_0 | fits | 32.6 GB | 256k | 4tok/sest. | 317 tok/s | – |
| Mistral Small 4 119B Mistral AI | UD-Q6_K | fits | 93.6 GB | 93k148k with q8 cache | 14tok/sest. | 1257 tok/s | – |
| Qwen 3.6 35B-A3B Alibaba Qwen | Q8_0 | fits | 35.4 GB | 256k | 26tok/sest. | 3116 tok/s | chat, coding |
| Qwen 3.5 27B Alibaba Qwen | Q8_0 | fits | 28.1 GB | 256k | 5tok/sest. | 364 tok/s | – |
| Gemma 4 26B-A4B | Q8_0 | fits | 26.2 GB | 256k | 19tok/sest. | 2375 tok/s | chat |
| GLM 4.5 Air Z.ai | Q6_K | fits | 94.5 GB | 16k29k with q8 cache | 5tok/sest. | 480 tok/s | – |
| Qwen 3.5 35B-A3B Alibaba Qwen | Q8_0 | fits | 35.4 GB | 256k | 26tok/sest. | 3116 tok/s | chat, coding |
| Qwen3 Next 80B-A3B Alibaba Qwen | Q8_0 | fits | 80.1 GB | 256k | 25tok/sest. | 3055 tok/s | chat, coding |
| GLM 4.7 Flash Z.ai | Q8_0 | fits | 30.9 GB | 198k | 15tok/sest. | 1744 tok/s | – |
| Devstral Small 2 24B Mistral AI | Q8_0 | fits | 25.4 GB | 384k | 5tok/sest. | 410 tok/s | – |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA | UD-Q6_K_XL | fits | 33.5 GB | 256k | 25tok/sest. | 3247 tok/s | chat, coding, long documents |
| gpt-oss 20B OpenAI | MXFP4 | fits | 12.3 GB | 128k | 32tok/sest. | 2256 tok/s | chat, coding |
| Kimi Linear 48B-A3B Moonshot AI | Q8_0 | fits | 49.6 GB | 1024k | 26tok/sest. | 2962 tok/s | chat, coding, long documents |
| Qwen3 Coder 30B-A3B Alibaba Qwen | Q8_0 | fits | 31.8 GB | 256k | 21tok/sest. | 2255 tok/s | chat |
| Qwen3 32B Alibaba Qwen | Q8_0 | fits | 35.3 GB | 40k | 4tok/sest. | 281 tok/s | – |
| Gemma 4 12B | Q8_0 | fits | 13.1 GB | 256k | 11tok/sest. | 827 tok/s | – |
| Qwen 3.5 9B Alibaba Qwen | Q8_0 | fits | 10.0 GB | 256k | 14tok/sest. | 1048 tok/s | – |
| Mistral Small 3.2 24B Mistral AI | Q8_0 | fits | 25.4 GB | 128k | 5tok/sest. | 410 tok/s | – |
| Gemma 3 27B | Q8_0 | fits | 28.6 GB | 128k | 5tok/sest. | 369 tok/s | – |
| DeepSeek R1 Distill Qwen 32B DeepSeek | Q8_0 | fits | 35.3 GB | 128k | 4tok/sest. | 293 tok/s | – |
| Llama 3.3 70B Meta | Q8_0 | fits | 73.2 GB | 79k128k with q8 cache | 2tok/sest. | 137 tok/s | – |
| Ministral 3 14B Mistral AI | Q8_0 | fits | 15.5 GB | 256k | 9tok/sest. | 680 tok/s | – |
| Qwen3 14B Alibaba Qwen | Q8_0 | fits | 16.7 GB | 40k | 8tok/sest. | 632 tok/s | – |
| DeepSeek R1 0528 Qwen3 8B DeepSeek | Q8_0 | fits | 10.1 GB | 128k | 14tok/sest. | 1106 tok/s | – |
| Qwen3 8B Alibaba Qwen | Q8_0 | fits | 10.1 GB | 40k | 14tok/sest. | 1106 tok/s | – |
| Phi-4 14B Microsoft | Q8_0 | fits | 16.9 GB | 16k | 8tok/sest. | 636 tok/s | – |
| Qwen 3.5 4B Alibaba Qwen | Q8_0 | fits | 5.3 GB | 256k | 29tok/sest. | 2109 tok/s | chat, coding, long documents |
| Ministral 3 8B Mistral AI | Q8_0 | fits | 10.3 GB | 256k | 14tok/sest. | 1033 tok/s | – |
| Granite 4.2 8B IBM | Q8_0 | fits | 10.8 GB | 128k | 13tok/sest. | 1026 tok/s | – |
| Gemma 4 E4B | Q8_0 | fits | 8.7 GB | 128k | 17tok/sest. | 1270 tok/s | chat |
| LFM2.5 8B-A1B Liquid AI | Q8_0 | fits | 9.6 GB | 125k | 58tok/sest. | 7411 tok/s | chat, coding |
| Llama 3.1 8B Meta | Q6_K | fits | 8.0 GB | 128k | 18tok/sest. | 1142 tok/s | chat |
| Qwen2.5 0.5B Alibaba Qwen | Q8_0 | fits | 1.4 GB | 32k | 223tok/sest. | 15510 tok/s | chat, coding, long documents |
| Qwen3.5 0.8B Alibaba Qwen | Q8_0 | fits | 1.7 GB | 256k | 151tok/sest. | 10671 tok/s | chat, coding, agents, long documents |
| Qwen3.5 2B Alibaba Qwen | Q5_K_M | fits | 2.3 GB | 256k | 91tok/sest. | 4377 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 1.5B DeepSeek | Q8_0 | fits | 2.8 GB | 128k | 66tok/sest. | 4882 tok/s | chat, coding, long documents |
| MiniCPM5 2B openbmb | Q8_0 | fits | 3.7 GB | 128k | 46tok/sest. | 3227 tok/s | chat, coding, agents, long documents |
| LFM2.5 2.6B Liquid AI | Q8_0 | fits | 4.0 GB | 128k | 42tok/sest. | 3248 tok/s | chat, coding |
| LFM2.5 VL 3B Liquid AI | Q8_0 | fits | 4.0 GB | 32k | 42tok/sest. | 2866 tok/s | chat, coding |
| granite vision 4.1 4b IBM | Q8_0 | fits | 4.8 GB | 128k | 33tok/sest. | 2150 tok/s | chat, coding |
| granite 4.2 3b IBM | Q8_0 | fits | 5.1 GB | 128k | 31tok/sest. | 2311 tok/s | chat, coding |
| LFM2.5 24B A2B Liquid AI | Q8_0 | fits | 25.2 GB | 128k | 30tok/sest. | 3892 tok/s | chat |
| Mellum2.1 12B A2.5B Thinking JetBrains | Q8_0 | fits | 13.0 GB | 128k | 30tok/sest. | 3571 tok/s | chat, coding, long documents |
| Kimi VL A3B Thinking 2506 Moonshot AI | Q8_0 | fits | 16.9 GB | 128k | 25tok/sest. | 3010 tok/s | chat, coding |
| Kimi VL A3B Moonshot AI | mlx-8bit | fits | 17.7 GB | 128k | 24tok/sest. | 3010 tok/s | chat |
| DeepSeek R1 Distill Qwen 7B DeepSeek | Q8_0 | fits | 8.8 GB | 128k | 16tok/sest. | 1232 tok/s | chat |
| GLM 4 9B 0414 Z.ai | mlx-5bit | fits | 8.8 GB | 32k | 16tok/sest. | 968 tok/s | chat |
| DeepSeek R1 Distill Llama 8B DeepSeek | Q8_0 | fits | 9.8 GB | 128k | 15tok/sest. | 1142 tok/s | – |
| Ministral 3 8B Reasoning 2512 Mistral AI | Q8_0 | fits | 10.3 GB | 256k | 14tok/sest. | 1033 tok/s | – |
| GLM 4.6V Flash Z.ai | Q8_0 | fits | 10.5 GB | 128k | 14tok/sest. | 893 tok/s | – |
| GLM 4.1V 9B Thinking Z.ai | mlx-8bit | fits | 12.1 GB | 64k | 12tok/sest. | 893 tok/s | – |
| GLM 4.5V Z.ai | Q4_K_M | fits | 61.5 GB | 64k | 9tok/sest. | 550 tok/s | – |
| DeepSeek R1 Distill Qwen 14B DeepSeek | Q8_0 | fits | 17.0 GB | 128k | 8tok/sest. | 619 tok/s | – |
| GLM 4.6V Z.ai | Q4_K_M | fits | 67.9 GB | 128k | 8tok/sest. | 550 tok/s | – |
| Devstral Small 2507 Mistral AI | Q8_0 | fits | 25.4 GB | 128k | 5tok/sest. | 417 tok/s | – |
| Magistral Small 2506 Mistral AI | Q8_0 | fits | 25.4 GB | 40k | 5tok/sest. | 417 tok/s | – |
| Magistral Small 2507 Mistral AI | Q8_0 | fits | 25.4 GB | 40k | 5tok/sest. | 417 tok/s | – |
| Magistral Small 2509 Mistral AI | Q8_0 | fits | 25.4 GB | 128k | 5tok/sest. | 410 tok/s | – |
| granite 4.2 30b IBM | Q8_0 | fits | 31.8 GB | 128k | 4tok/sest. | 331 tok/s | – |
| DeepSeek R1 Distill Llama 70B DeepSeek | Q8_0 | fits | 73.2 GB | 79k128k with q8 cache | 2tok/sest. | 137 tok/s | – |
| Kimi Dev 72B Moonshot AI | Q8_0 | fits | 75.3 GB | 73k128k with q8 cache | 2tok/sest. | 133 tok/s | – |
| Devstral 2 123B Instruct 2512 Mistral AI | Q5_K_M | fits | 85.8 GB | 37k68k with q8 cache | 2tok/sest. | 78 tok/s | – |
| Mistral Medium 3.5 128B Mistral AI | Q5_K_M | fits | 85.8 GB | 37k68k with q8 cache | 2tok/sest. | 76 tok/s | – |
| GLM 5.3 Flash Z.ai | UD-Q2_K_XL | won't run | 102.3 GB | – | – | – | – |
| DeepSeek R1 DeepSeek | Q4_K_M | won't run | 378.0 GB | – | – | – | – |
| DeepSeek R1 0528 DeepSeek | Q4_K_M | won't run | 378.5 GB | – | – | – | – |
| DeepSeek R1 Zero DeepSeek | Q4_K_M | won't run | 378.0 GB | – | – | – | – |
| DeepSeek V3 0324 DeepSeek | Q4_K_M | won't run | 378.0 GB | – | – | – | – |
| DeepSeek V3.1 DeepSeek | Q4_K_M | won't run | 378.9 GB | – | – | – | – |
| DeepSeek V3.1 Terminus DeepSeek | Q4_K_M | won't run | 378.9 GB | – | – | – | – |
| DeepSeek V3.2 DeepSeek | Q4_K_M | won't run | 378.9 GB | – | – | – | – |
| DeepSeek V4 Flash 0731 DeepSeek | MXFP4 | won't run | 147.2 GB | – | – | – | – |
| DeepSeek V4 Flash Vision Exp DeepSeek | MXFP4 | won't run | 155.9 GB | – | – | – | – |
| GLM 4.6 Z.ai | Q4_K_M | won't run | 204.5 GB | – | – | – | – |
| GLM 4.7 Z.ai | Q4_K_M | won't run | 205.3 GB | – | – | – | – |
| GLM 5.1 Z.ai | MXFP4 | won't run | 421.5 GB | – | – | – | – |
| GLM 5.2 Z.ai | UD-Q4_K_M | won't run | 435.4 GB | – | – | – | – |
| GLM 5.3 Z.ai | Q8_0 | won't run | 747.9 GB | – | – | – | – |
| Kimi K2 Moonshot AI | Q4_K_M | won't run | 579.5 GB | – | – | – | – |
| Kimi K2 Instruct 0905 Moonshot AI | Q4_K_M | won't run | 580.0 GB | – | – | – | – |
| Kimi K2 Thinking Moonshot AI | Q4_K_M | won't run | 580.0 GB | – | – | – | – |
| Kimi K2.5 Moonshot AI | Q4_K_M | won't run | 580.0 GB | – | – | – | – |
| MiniMax M2 MiniMax | Q4_K_M | won't run | 131.6 GB | – | – | – | – |
| MiniMax M2.1 MiniMax | Q4_K_M | won't run | 131.6 GB | – | – | – | – |
| MiniMax M2.5 MiniMax | MXFP4 | won't run | 118.1 GB | – | – | – | – |
| MiniMax M3 MiniMax | UD-Q4_K_M | won't run | 248.5 GB | – | – | – | – |
Specifications
From the manufacturer's published specifications; sources are linked where we have them.
Compare with
Questions
- How much memory does the Beelink GTR9 Pro Ryzen AI Max+ 395 128GB have for AI models?
- The Beelink GTR9 Pro Ryzen AI Max+ 395 128GB has 128 GB of unified memory, of which the GPU can use 96 GB by default, with 256 GB/s of memory bandwidth.
- What is the best model the Beelink GTR9 Pro Ryzen AI Max+ 395 128GB can run locally?
- DeepSeek V4 Flash (UD-Q2_K_XL) fits fully in its memory at 8k context and generates at about 11 tok/s, an estimate from its memory bandwidth. 72 of the 113 models we track fit at that context.
- Can the Beelink GTR9 Pro Ryzen AI Max+ 395 128GB run GLM 5.3 Flash?
- No: even at UD-Q2_K_XL, GLM 5.3 Flash needs about 102 GB at 8k context, more than the Beelink GTR9 Pro Ryzen AI Max+ 395 128GB and system RAM can hold.
- How much does the Beelink GTR9 Pro Ryzen AI Max+ 395 128GB cost in Germany?
- We have no current price for the Beelink GTR9 Pro Ryzen AI Max+ 395 128GB in Germany.
Speeds on this page are for 8k context. See the best hardware for local AI at every budget.