What hardware do I need?
Hardware that delivers the experience you want — not just loads the model
Pick the model and describe the workload. We rate every machine in the database for that exact workload and group them by whether they meet your target.
1 · Model
2 · Use case
Agentic coding
Chat & writing
Coding
Documents & research
Multimodal
Automation & serving
3 · Tool & runtime
Tool: OpenCode
4 · Workload
Project size, agent behaviour, other apps
5 · Desired experience
6 · Budget & vendors (optional)
Prices are approximate launch prices for the whole system.
Gemma 4 31B · Agentic coding / medium repository
Target: comfortable · 14 machines meet it
Meets your comfortable target (14)
Cheapest first.
Desktop PC · GeForce RTX 3090 · 32 GB RAM
24 GB VRAM (936 GB/s) + 32 GB RAM
- Quant / runtime
- Q4 · Ollama
- Generation
- 25–35 tok/s
- Per step
- ≈15 s
- Headroom
- 1.3 GB
Desktop PC · Radeon RX 7900 XTX · 32 GB RAM
24 GB VRAM (960 GB/s) + 32 GB RAM
- Quant / runtime
- Q4 · Lemonade
- Generation
- 30–40 tok/s
- Per step
- ≈15 s
- Headroom
- 1.3 GB
Desktop PC · GeForce RTX 3090 · 64 GB RAM
24 GB VRAM (936 GB/s) + 64 GB RAM
- Quant / runtime
- Q4 · Ollama
- Generation
- 25–35 tok/s
- Per step
- ≈15 s
- Headroom
- 1.3 GB
Desktop PC · Radeon RX 7900 XTX · 64 GB RAM
24 GB VRAM (960 GB/s) + 64 GB RAM
- Quant / runtime
- Q4 · Lemonade
- Generation
- 30–40 tok/s
- Per step
- ≈15 s
- Headroom
- 1.3 GB
Desktop PC · GeForce RTX 4090 · 32 GB RAM
24 GB VRAM (1008 GB/s) + 32 GB RAM
- Quant / runtime
- Q4 · Ollama
- Generation
- 30–40 tok/s
- Per step
- ≈10 s
- Headroom
- 1.3 GB
Desktop PC · GeForce RTX 4090 · 64 GB RAM
24 GB VRAM (1008 GB/s) + 64 GB RAM
- Quant / runtime
- Q4 · Ollama
- Generation
- 30–40 tok/s
- Per step
- ≈10 s
- Headroom
- 1.3 GB
Desktop PC · 2× GeForce RTX 3090 (48 GB) · 128 GB RAM
48 GB VRAM (936 GB/s) + 128 GB RAM
- Quant / runtime
- Q4 · Ollama
- Generation
- 25–40 tok/s
- Per step
- ≈15 s
- Headroom
- 25 GB
Desktop PC · GeForce RTX 5090 · 32 GB RAM
32 GB VRAM (1792 GB/s) + 32 GB RAM
- Quant / runtime
- Q6 · Ollama
- Generation
- 35–45 tok/s
- Per step
- ≈10 s
- Headroom
- 2.1 GB
Desktop PC · GeForce RTX 5090 · 64 GB RAM
32 GB VRAM (1792 GB/s) + 64 GB RAM
- Quant / runtime
- Q6 · Ollama
- Generation
- 35–45 tok/s
- Per step
- ≈10 s
- Headroom
- 2.1 GB
Meets the acceptable target (29)
Usable, with noticeable compromises for this workload.
MacBook Pro M5 Max 48 GB
48 GB unified memory · 614 GB/s
- Quant / runtime
- Q4 · LM Studio
- Generation
- 16–25 tok/s
- Per step
- ≈25 s
- Headroom
- 15 GB
MacBook Pro M5 Max 64 GB
64 GB unified memory · 614 GB/s
- Quant / runtime
- Q4 · LM Studio
- Generation
- 16–25 tok/s
- Per step
- ≈25 s
- Headroom
- 30 GB
MacBook Pro M5 Max 128 GB
128 GB unified memory · 614 GB/s
- Quant / runtime
- Q4 · LM Studio
- Generation
- 16–25 tok/s
- Per step
- ≈25 s
- Headroom
- 94 GB
Mac Studio M5 Max 48 GB
48 GB unified memory · 614 GB/s
- Quant / runtime
- Q4 · LM Studio
- Generation
- 16–25 tok/s
- Per step
- ≈25 s
- Headroom
- 15 GB
Mac Studio M5 Max 64 GB
64 GB unified memory · 614 GB/s
- Quant / runtime
- Q4 · LM Studio
- Generation
- 16–25 tok/s
- Per step
- ≈25 s
- Headroom
- 30 GB
Mac Studio M5 Max 128 GB
128 GB unified memory · 614 GB/s
- Quant / runtime
- Q4 · LM Studio
- Generation
- 16–25 tok/s
- Per step
- ≈25 s
- Headroom
- 94 GB
Desktop PC · Radeon AI PRO R9700 · 32 GB RAM
32 GB VRAM (640 GB/s) + 32 GB RAM
- Quant / runtime
- Q6 · Lemonade
- Generation
- 17–25 tok/s
- Per step
- ≈20 s
- Headroom
- 2.1 GB
Desktop PC · Radeon AI PRO R9700 · 64 GB RAM
32 GB VRAM (640 GB/s) + 64 GB RAM
- Quant / runtime
- Q6 · Lemonade
- Generation
- 17–25 tok/s
- Per step
- ≈20 s
- Headroom
- 2.1 GB
Mac Studio M3 Ultra 96 GB
96 GB unified memory · 819 GB/s
- Quant / runtime
- Q4 · LM Studio
- Generation
- 17–25 tok/s
- Per step
- ≈30 s
- Headroom
- 62 GB
Can run, but below your target (53)
The model loads, but the experience falls short for this workload.
Cannot run (32)
Not enough usable memory, or no compatible runtime.