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.
Nemotron 3.5 Super VL 120B-A12B · Agentic coding / medium repository
Target: comfortable · 13 machines meet it
Meets your comfortable target (13)
Cheapest first.
Mac Studio M4 Max 128 GB
128 GB unified memory · 546 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 35–45 tok/s
- Per step
- ≈20 s
- Headroom
- 53 GB
MacBook Pro M4 Max 128 GB
128 GB unified memory · 546 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 35–45 tok/s
- Per step
- ≈20 s
- Headroom
- 53 GB
Mac Studio M5 Ultra 128 GB
128 GB unified memory · 1228 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 45–75 tok/s
- Per step
- ≈10 s
- Headroom
- 53 GB
Mac Studio M3 Ultra 96 GB
96 GB unified memory · 819 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 35–45 tok/s
- Per step
- ≈15 s
- Headroom
- 21 GB
Mac Studio M1 Ultra 128 GB
128 GB unified memory · 800 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 40–50 tok/s
- Per step
- ≈15 s
- Headroom
- 53 GB
Mac Studio M2 Ultra 128 GB
128 GB unified memory · 800 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 35–45 tok/s
- Per step
- ≈15 s
- Headroom
- 53 GB
Mac Studio M5 Ultra 256 GB
256 GB unified memory · 1228 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 45–75 tok/s
- Per step
- ≈10 s
- Headroom
- 181 GB
Mac Studio M3 Ultra 256 GB
256 GB unified memory · 819 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 35–45 tok/s
- Per step
- ≈15 s
- Headroom
- 181 GB
Mac Studio M2 Ultra 192 GB
192 GB unified memory · 800 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 35–45 tok/s
- Per step
- ≈15 s
- Headroom
- 117 GB
Meets the acceptable target (7)
Usable, with noticeable compromises for this workload.
MacBook Pro M2 Max 96 GB
96 GB unified memory · 400 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 25–35 tok/s
- Per step
- ≈25 s
- Headroom
- 21 GB
MacBook Pro M3 Max 128 GB
128 GB unified memory · 400 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 25–35 tok/s
- Per step
- ≈25 s
- Headroom
- 53 GB
MacBook Pro M5 Max 128 GB
128 GB unified memory · 614 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 30–50 tok/s
- Per step
- ≈20 s
- Headroom
- 53 GB
Mac Studio M2 Max 96 GB
96 GB unified memory · 400 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 25–35 tok/s
- Per step
- ≈25 s
- Headroom
- 21 GB
Mac Studio M5 Max 128 GB
128 GB unified memory · 614 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 30–50 tok/s
- Per step
- ≈20 s
- Headroom
- 53 GB
NVIDIA DGX Spark (GB10) · 128 GB
128 GB unified memory · 273 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 20–30 tok/s
- Per step
- ≈25 s
- Headroom
- 55 GB
MacBook Pro M3 Max 96 GB
96 GB unified memory · 300 GB/s
- Quant / runtime
- MXFP4 · vLLM
- Generation
- 20–25 tok/s
- Per step
- ≈30 s
- Headroom
- 21 GB
Can run, but below your target (0)
The model loads, but the experience falls short for this workload.
Cannot run (108)
Not enough usable memory, or no compatible runtime.