Editing a copy of RTX 5090 32GB
Choose parts or a complete machine. Compatibility, the total and what it runs update as you go, and nothing is saved until you share it.
Compatibility: No problems found.Est. load 820 W of 1200 W · spikes to 1165 W
Best fit:Qwen 3.8 27BQ8_0at 32k, 35 tok/sest.
| Component | Selection | Price | Where | Actions |
|---|---|---|---|---|
| GPU | NVIDIA GeForce RTX 5090 32 GB · 1792 GB/s · 575 W | £5,000 | Buy | |
| GPU 2optional | ||||
| CPU | AMD Ryzen 9 9950X AM5 · 16 cores · 24 PCIe lanes | £442 | Buy | |
| Motherboard | MSI MPG X870E Carbon WiFi AM5 · DDR5 · ATX | £441 | Buy | |
| Memory | Kingston Fury Beast 96GB (2×48GB) DDR5-6400 96 GB · DDR5-6400 | Not sold here | – | |
| Storage | WD Black SN850X 4TB 4000 GB NVME | £420 | Buy | |
| Power supply | Seasonic Vertex GX-1200 1200W 1200 W · ATX 3 | Not sold here | – | |
| Case | Lian Li O11 Dynamic EVO GPUs up to 426 mm | Not sold here | – | |
| Total + 3 without a price | £6,304 | Prices checked 32 min ago | ||
Notes
- Note: Check the power connectors on the card you buy: this power supply has 1 12V-2x6 and 3 8-pin PCIe connectors.
- Note: Lian Li O11 Dynamic EVO fits graphics cards up to 426 mm: check the length of the card you buy.
What this build runs
The most precise quant that fits, for every model in the catalogue. Estimates, not measurements.
| Model | Quant | Fits | Memory | Max context | Decode @32k | Prefill | Good for |
|---|---|---|---|---|---|---|---|
| Qwen 3.8 27B Alibaba Qwen | Q8_0 | fits | 30.1 GB | 51k89k with q8 cache | 35tok/sest. | 1517 tok/s | chat, coding, agents |
| Qwen 3.6 27B Alibaba Qwen | Q8_0 | fits | 29.7 GB | 57k100k with q8 cache | 36tok/sest. | 1517 tok/s | chat, coding, agents |
| Gemma 4 31B | Q6_K | fits | 27.6 GB | 76k143k with q8 cache | 39tok/sest. | 1351 tok/s | chat, coding, agents |
| Qwen 3.6 35B-A3B Alibaba Qwen | Q6_K | fits | 30.4 GB | 71k112k with q8 cache | 201tok/sest. | 10830 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 27B Alibaba Qwen | Q8_0 | fits | 29.7 GB | 57k100k with q8 cache | 36tok/sest. | 1517 tok/s | chat, coding, agents |
| Gemma 4 26B-A4B | Q8_0 | fits | 26.7 GB | 216k256k with q8 cache | 132tok/sest. | 9635 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 35B-A3B Alibaba Qwen | Q6_K | fits | 28.4 GB | 148k232k with q8 cache | 212tok/sest. | 10830 tok/s | chat, coding, agents, long documents |
| GLM 4.7 Flash Z.ai | Q6_K | fits | 25.5 GB | 133k198k with q8 cache | 114tok/sest. | 5654 tok/s | chat, coding, agents, long documents |
| Devstral Small 2 24B Mistral AI | Q8_0 | fits | 29.2 GB | 45k82k with q8 cache | 37tok/sest. | 1601 tok/s | chat, coding, agents |
| Kimi Linear 48B-A3B Moonshot AI | Q4_K_M | fits | 29.2 GB | 196k267k with q8 cache | 311tok/sest. | 9333 tok/s | chat, coding, long documents |
£6,304
No problems found