Dual used RTX 3090 48GB
£3,988
Parts
Compatibility: No problems found.Est. load 895 W of 1200 W · spikes to 1315 W
| Component | Selection | Price | Where | Buy |
|---|---|---|---|---|
| GPU | NVIDIA GeForce RTX 3090 24 GB · 936 GB/s · 350 W | £1,330 | Amazon | Buy |
| GPU 2 | NVIDIA GeForce RTX 3090 24 GB · 936 GB/s · 350 W | £1,330 | Amazon | Buy |
| CPU | AMD Ryzen 9 9900X AM5 · 12 cores · 24 PCIe lanes | £309 | eBay | Buy |
| Motherboard | ASUS ProArt X870E-Creator WiFi AM5 · DDR5 · ATX | £450 | Amazon | Buy |
| Memory | Corsair Vengeance 128GB (4×32GB) DDR5-5200 128 GB · DDR5-5200 | Not sold now | – | |
| Power supply | Seasonic Vertex GX-1200 1200W 1200 W · ATX 3 | £136 | eBay (used) | Buy |
| Case | Fractal Design Torrent GPUs up to 461 mm | £183 | eBay | Buy |
| Storage | Samsung 990 Pro 2TB 2000 GB NVME | £250 | eBay | Buy |
| Total + 1 without a price | £3,988 | |||
- Note: Check the power connectors on the card you buy: this power supply has 1 12V-2x6 and 3 8-pin PCIe connectors.
- Note: The second x16 slot is 3 slots below the first: choose cards no thicker than 3 slots, or the top card covers it.
- Note: Fractal Design Torrent fits graphics cards up to 461 mm: check the length of the card you buy.
What this build runs
The most precise quant that fits, at 128k context. Estimates, not measurements.
| Model | Quant | Fits | Memory | Max context | Decode @128k | Prefill | Good for |
|---|---|---|---|---|---|---|---|
| Qwen 3.8 27B Alibaba Qwen | Q8_0 | fits | 37.0 GB | 256k | 15tok/sest. | 393 tok/s | chat, long documents |
| Qwen 3.5 122B-A10B Alibaba Qwen | UD-Q2_K_XL | fits | 44.0 GB | 224k256k with q8 cache | 52tok/sest. | 698 tok/s | chat, coding, agents, long documents |
| Qwen 3.6 27B Alibaba Qwen | Q8_0 | fits | 36.6 GB | 256k | 16tok/sest. | 393 tok/s | chat, long documents |
| Gemma 4 31B | Q8_0 | fits | 43.0 GB | 172k256k with q8 cache | 13tok/sest. | 373 tok/s | chat |
| Mistral Small 4 119B Mistral AI | UD-Q2_K_XL | fits | 42.1 GB | 296k471k with q8 cache | 65tok/sest. | 605 tok/s | chat, coding, agents, long documents |
| Qwen 3.6 35B-A3B Alibaba Qwen | Q8_0 | fits | 38.8 GB | 256k | 60tok/sest. | 1909 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 27B Alibaba Qwen | Q8_0 | fits | 36.6 GB | 256k | 16tok/sest. | 393 tok/s | chat, long documents |
| Gemma 4 26B-A4B | Q8_0 | fits | 29.6 GB | 256k | 49tok/sest. | 2317 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 35B-A3B Alibaba Qwen | Q8_0 | fits | 38.8 GB | 256k | 60tok/sest. | 1909 tok/s | chat, coding, agents, long documents |
| GLM 4.7 Flash Z.ai | Q8_0 | fits | 38.1 GB | 198k | 28tok/sest. | 900 tok/s | chat, coding, agents, long documents |
| Devstral Small 2 24B Mistral AI | Q8_0 | fits | 45.2 GB | 137k251k with q8 cache | 13tok/sest. | 351 tok/s | chat |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA | UD-Q6_K_XL | fits | 35.2 GB | 256k | 83tok/sest. | 2568 tok/s | chat, coding, agents, long documents |
| Kimi Linear 48B-A3B Moonshot AI | Q6_K | fits | 40.6 GB | 588k802k with q8 cache | 98tok/sest. | 1432 tok/s | chat, coding, long documents |
| gpt-oss 20B OpenAI | MXFP4 | fits | 16.1 GB | 128k | 62tok/sest. | 1500 tok/s | chat, coding, agents, long documents |
| Qwen3 Coder 30B-A3B Alibaba Qwen | Q8_0 | fits | 44.1 GB | 155k256k with q8 cache | 22tok/sest. | 555 tok/s | chat, coding, agents, long documents |
| Gemma 4 12B | Q8_0 | fits | 16.0 GB | 256k | 38tok/sest. | 963 tok/s | chat, coding, long documents |
| Qwen 3.5 9B Alibaba Qwen | Q8_0 | fits | 14.8 GB | 256k | 42tok/sest. | 1145 tok/s | chat, coding, long documents |
| Gemma 3 27B | Q8_0 | fits | 39.0 GB | 128k | 15tok/sest. | 481 tok/s | chat |
| Mistral Small 3.2 24B Mistral AI | Q8_0 | fits | 45.2 GB | 128k | 13tok/sest. | 351 tok/s | chat |
| Ministral 3 14B Mistral AI | Q8_0 | fits | 35.2 GB | 199k256k with q8 cache | 16tok/sest. | 451 tok/s | chat, coding, long documents |
| Qwen3 14B Alibaba Qwen | Q8_0 | fits | 36.5 GB | 40k | 16tok/sest. | 384 tok/s | chat, coding, long documents |
| DeepSeek R1 0528 Qwen3 8B DeepSeek | Q8_0 | fits | 28.0 GB | 128k | 21tok/sest. | 581 tok/s | chat, coding, long documents |
| Qwen3 8B Alibaba Qwen | Q8_0 | fits | 28.0 GB | 40k | 21tok/sest. | 581 tok/s | chat, coding, long documents |
| Phi-4 14B Microsoft | Q8_0 | fits | 41.4 GB | 16k | 14tok/sest. | 385 tok/s | chat, coding |
| Qwen 3.5 4B Alibaba Qwen | Q8_0 | fits | 10.1 GB | 256k | 66tok/sest. | 1784 tok/s | chat, coding, long documents |
| Ministral 3 8B Mistral AI | Q8_0 | fits | 27.3 GB | 256k | 21tok/sest. | 588 tok/s | chat, coding, long documents |
| Granite 4.2 8B IBM | Q8_0 | fits | 30.6 GB | 128k | 19tok/sest. | 528 tok/s | chat, coding, agents, long documents |
| Gemma 4 E4B | Q8_0 | fits | 13.0 GB | 128k | 48tok/sest. | 1606 tok/s | chat, coding, long documents |
| LFM2.5 8B-A1B Liquid AI | Q8_0 | fits | 16.2 GB | 125k | 46tok/sest. | 2147 tok/s | chat, coding, long documents |
| Llama 3.1 8B Meta | Q6_K | fits | 24.0 GB | 128k | 24tok/sest. | 634 tok/s | chat, coding, long documents |
| Qwen2.5 0.5B Alibaba Qwen | Q8_0 | fits | 3.8 GB | 32k | 271tok/sest. | 4823 tok/s | chat, coding, long documents |
| Qwen3.5 0.8B Alibaba Qwen | Q8_0 | fits | 4.1 GB | 256k | 238tok/sest. | 6431 tok/s | chat, coding, agents, long documents |
| Qwen3.5 2B Alibaba Qwen | Q5_K_M | fits | 4.7 GB | 256k | 189tok/sest. | 4112 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 1.5B DeepSeek | Q8_0 | fits | 7.1 GB | 128k | 103tok/sest. | 2155 tok/s | chat, coding, long documents |
| Mellum2.1 12B A2.5B Thinking JetBrains | Q8_0 | fits | 15.7 GB | 128k | 76tok/sest. | 2483 tok/s | chat, coding, agents, long documents |
| MiniCPM5 2B openbmb | Q8_0 | fits | 9.6 GB | 128k | 70tok/sest. | 1158 tok/s | chat, coding, agents, long documents |
| LFM2.5 2.6B Liquid AI | Q8_0 | fits | 12.0 GB | 128k | 53tok/sest. | 1486 tok/s | chat, coding, agents, long documents |
| LFM2.5 VL 3B Liquid AI | Q8_0 | fits | 12.0 GB | 32k | 53tok/sest. | 1429 tok/s | chat, coding, agents, long documents |
| Kimi VL A3B Moonshot AI | mlx-8bit | fits | 22.2 GB | 128k | 48tok/sest. | 1559 tok/s | chat, coding, agents, long documents |
| granite vision 4.1 4b IBM | Q8_0 | fits | 15.2 GB | 128k | 40tok/sest. | 917 tok/s | chat, coding, long documents |
| granite 4.2 3b IBM | Q8_0 | fits | 15.5 GB | 128k | 40tok/sest. | 935 tok/s | chat, coding, agents, long documents |
| GLM 4.6V Flash Z.ai | Q8_0 | fits | 16.2 GB | 128k | 38tok/sest. | 503 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 7B DeepSeek | Q8_0 | fits | 16.4 GB | 128k | 37tok/sest. | 769 tok/s | chat, coding, long documents |
| GLM 4.1V 9B Thinking Z.ai | mlx-8bit | fits | 17.8 GB | 64k | 34tok/sest. | 503 tok/s | chat, coding, long documents |
| LFM2.5 24B A2B Liquid AI | Q8_0 | fits | 35.6 GB | 128k | 27tok/sest. | 1254 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Llama 8B DeepSeek | Q8_0 | fits | 25.8 GB | 128k | 23tok/sest. | 634 tok/s | chat, coding, long documents |
| Ministral 3 8B Reasoning 2512 Mistral AI | Q8_0 | fits | 27.3 GB | 256k | 21tok/sest. | 588 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 14B DeepSeek | Q8_0 | fits | 40.5 GB | 128k | 14tok/sest. | 340 tok/s | chat, coding |
| Devstral Small 2507 Mistral AI | Q8_0 | fits | 45.2 GB | 128k | 13tok/sest. | 355 tok/s | chat |
| Magistral Small 2506 Mistral AI | Q8_0 | fits | 45.2 GB | 40k | 13tok/sest. | 355 tok/s | chat |
| Magistral Small 2507 Mistral AI | Q8_0 | fits | 45.2 GB | 40k | 13tok/sest. | 355 tok/s | chat |
| Magistral Small 2509 Mistral AI | Q8_0 | fits | 45.2 GB | 128k | 13tok/sest. | 351 tok/s | chat |
| DeepSeek V4 Flash DeepSeek | UD-Q2_K_XL | offload | 102.8 GB | – | 3tok/sest. | 52 tok/s | – |
| GLM 5.3 Flash Z.ai | UD-Q2_K_XL | offload | 104.6 GB | – | 5tok/sest. | 87 tok/s | – |
| Qwen 3.8 Flash Next Alibaba Qwen | UD-Q2_K_XL | offload | 78.4 GB | – | 2tok/sest. | 169 tok/s | – |
| MiniMax M2.7 MiniMax | UD-Q2_K_XL | offload | 103.0 GB | – | 1tok/sest. | 175 tok/s | – |
| Qwen3 Coder Next Alibaba Qwen | Q4_K_M | offload | 50.1 GB | 14k23k with q8 cache | 36tok/sest. | 1485 tok/s | chat, coding, agents, long documents |
| Nemotron 3 Super 120B-A12B NVIDIA | UD-Q2_K_XL | offload | 53.9 GB | – | 22tok/sest. | 900 tok/s | chat, coding, long documents |
| gpt-oss 120B OpenAI | Q4_K_M | offload | 64.8 GB | – | 11tok/sest. | 857 tok/s | chat, coding |
| GLM 4.5 Air Z.ai | Q2_K | offload | 67.1 GB | 18k34k with q8 cache | 3tok/sest. | 143 tok/s | chat |
| Qwen3 Next 80B-A3B Alibaba Qwen | Q4_K_M | offload | 50.1 GB | 15k24k with q8 cache | 40tok/sest. | 1623 tok/s | chat, coding, agents, long documents |
| Qwen3 32B Alibaba Qwen | Q4_K_M | offload | 52.3 GB | 40k | 5tok/sest. | 147 tok/s | chat, coding |
| DeepSeek R1 Distill Qwen 32B DeepSeek | Q4_K_M | offload | 52.3 GB | 106k128k with q8 cache | 5tok/sest. | 197 tok/s | chat, coding |
| Llama 3.3 70B Meta | Q4_K_M | offload | 81.5 GB | 19k35k with q8 cache | 1tok/sest. | 75 tok/s | – |
| granite 4.2 30b IBM | Q4_K_M | offload | 50.6 GB | 112k128k with q8 cache | 6tok/sest. | 239 tok/s | chat, coding |
| GLM 4.5V Z.ai | Q4_K_M | offload | 84.1 GB | – | 2tok/sest. | 132 tok/s | – |
| GLM 4.6V Z.ai | Q4_K_M | offload | 90.5 GB | – | 2tok/sest. | 127 tok/s | – |
| DeepSeek V4 Flash 0731 DeepSeek | MXFP4 | offload | 158.2 GB | – | 1tok/sest. | 42 tok/s | – |
| DeepSeek V4 Flash Vision Exp DeepSeek | MXFP4 | offload | 167.0 GB | – | 1tok/sest. | 41 tok/s | – |
| DeepSeek R1 Distill Llama 70B DeepSeek | Q4_K_M | offload | 81.5 GB | 19k35k with q8 cache | 1tok/sest. | 75 tok/s | – |
| MiniMax M2.5 MiniMax | MXFP4 | offload | 148.1 GB | – | 1tok/sest. | 153 tok/s | – |
| MiniMax M2 MiniMax | Q4_K_M | offload | 161.7 GB | – | 1tok/sest. | 149 tok/s | – |
| MiniMax M2.1 MiniMax | Q4_K_M | offload | 161.7 GB | – | 1tok/sest. | 149 tok/s | – |
| Mistral Medium 3.5 128B Mistral AI | Q4_K_M | offload | 115.6 GB | – | 1tok/sest. | 38 tok/s | – |
| DeepSeek R1 DeepSeek | Q4_K_M | won't run | 387.1 GB | – | – | – | – |
| DeepSeek R1 0528 DeepSeek | Q4_K_M | won't run | 387.6 GB | – | – | – | – |
| DeepSeek V3 0324 DeepSeek | Q4_K_M | won't run | 387.1 GB | – | – | – | – |
| DeepSeek V3.1 DeepSeek | Q4_K_M | won't run | 388.0 GB | – | – | – | – |
| DeepSeek V3.2 DeepSeek | Q4_K_M | won't run | 388.0 GB | – | – | – | – |
| Devstral 2 123B Instruct 2512 Mistral AI | Q8_0 | won't run | 169.6 GB | – | – | – | – |
| GLM 4.7 Z.ai | Q4_K_M | won't run | 249.4 GB | – | – | – | – |
| GLM 5.1 Z.ai | MXFP4 | won't run | 432.8 GB | – | – | – | – |
| GLM 5.2 Z.ai | UD-Q4_K_M | won't run | 446.7 GB | – | – | – | – |
| GLM 5.3 Z.ai | Q8_0 | won't run | 759.1 GB | – | – | – | – |
| Kimi K2 Moonshot AI | Q4_K_M | won't run | 588.6 GB | – | – | – | – |
| Kimi K2 Instruct 0905 Moonshot AI | Q4_K_M | won't run | 589.0 GB | – | – | – | – |
| Kimi K2 Thinking Moonshot AI | Q4_K_M | won't run | 589.0 GB | – | – | – | – |
| Kimi K2.5 Moonshot AI | Q4_K_M | won't run | 589.0 GB | – | – | – | – |
| MiniMax M3 MiniMax | UD-Q4_K_M | won't run | 263.6 GB | – | – | – | – |