Apple silicon
Apple Mac mini (M5 Pro)
- Memory
- 64 GB
- Bandwidth
- 307 GB/s
Largest popular model at 4-bit: Qwen3.6 35B A3B
Up to 64 GB at 307 GB/s in a quiet mini PC: 27–35B models at a comfortable pace.
from $1,699 (launch price)Recommended
NVIDIA · NVIDIA GPUs · 2025
NVIDIA RTX PRO 5000 Blackwell
Illustration
For advanced users
48 GB in a 300 W blower card: 70B dense models on one quiet-ish GPU.
- Memory
- 48 GB
- Bandwidth
- 1.34 TB/s
- Power
- 300 W
- Largest popular model at 4-bit
- Qwen3.6 35B A3B
from $4,500 (typical price)
Researched from manufacturer data and public sources; not tested by us.· Last reviewed Oct 11, 2026
What it can run
EstimatePopular models at Q4_K_M with a 32K-token context. Memory needed from real model files; speed is the theoretical ceiling.
Change model size or context
Model weightsContext cache (KV)RuntimeUsable limit (colour = verdict)
| Model | Verdict | Memory | Speed ceiling |
|---|---|---|---|
| Qwen3 1.7B | Runs | ≤ 1,211 tok/s | |
| Qwen3.5 9B | Runs | ≤ 237 tok/s | |
| Gemma 4 12B | Runs | ≤ 189 tok/s | |
| gpt-oss-20b | Runs | ≤ 671 tok/s | |
| Qwen3.6 27B | Runs | ≤ 80 tok/s | |
| Qwen3.6 35B A3B | Runs | ≤ 729 tok/s | |
| Llama 3.3 70B | Not enough memory | — | |
| gpt-oss-120b | Partly in system RAM | — | |
| Qwen3.5 122B A10B | Partly in system RAM | — | |
| Qwen3.5 397B A17B | Partly in system RAM | — |
Theoretical maximum for generating text: memory bandwidth divided by the bytes read per token. Real runs usually reach about half to three quarters of it.
Where to buy
Prices are editorial references (launch price and a typical street price on the date shown), not live prices. In 2026 the memory shortage has pushed many prices well above launch.
- RTX PRO 5000 Blackwell 48 GB48 GBTypical price: ~$4,500 (checked Oct 1, 2026)
- Check price Amazon.com (opens in a new tab)Amazon.comCheck availability
Other regions (3)
- Check price Amazon.co.uk (opens in a new tab)Amazon.co.ukUnited Kingdom · Check availability · Search results
- Check price Amazon.es (opens in a new tab)Amazon.esSpain · Check availability
- Check price Amazon.de (opens in a new tab)Amazon.deGermany · Check availability · Search results
Prices are shown only if updated in the last 24 hours and may change; the retailer’s price at checkout applies. We may earn a commission from these links at no cost to you. Affiliate disclosure
Overview
Editorial opinionBest for: Workstations and servers that need 48 GB without two consumer cards.
Editorial opinionWould we buy it today?
Good for a compact 48 GB server; two used RTX 3090s are cheaper if you have the room.
Pros and cons
Editorial opinionPros
- 48 GB on one card
- 300 W, blower cooler that stacks well
Cons
- Expensive per GB compared with used RTX 3090s
- Blower fans are loud under load
Specifications
Factual data- Manufacturer
- NVIDIA
- Chip
- GB202
- Architecture
- Blackwell
- Type
- Graphics card
- Form factor
- PCIe card
- Released
- 2025
- Availability
- On sale
- Memory options
- 48 GB
- Memory type
- GDDR7 ECC
- Memory bus
- 384-bit
- Memory bandwidth
- 1.34 TB/s
- Usable for models (default)
- 48 GB / 48 GB
- Board / max power
- 300 W
- Interface
- PCIe 5.0 x16
- Slots
- 2
- Power connector
- 16-pin 12V-2x6
- Compute stack
- CUDA
- Models
- Any model that fits
- Several units for one model
- Yes
Source: Manufacturer specifications
Software support
Editorial opinionHow well the popular local-AI tools run on it today (editorial assessment).
- llama.cppExcellent
- OllamaExcellent
- LM StudioExcellent
- vLLMExcellent
- MLXNot supported
Measured speeds
Factual dataMeasured results with their source. We never invent speeds.
No measured results yet. The table above shows the theoretical ceiling instead.
Alternatives
Apple silicon
Apple Mac Studio (M5 Ultra)
- Memory
- 512 GB
- Bandwidth
- 1.2 TB/s
Largest popular model at 4-bit: Qwen3.5 397B A17B
Up to 512 GB at 1.2 TB/s: the only single box that holds DeepSeek R1 or Qwen3.5 397B at 4-bit.
from $5,499 (launch price)Recommended
DGX Spark / GB10
NVIDIA GB10 Grace Blackwell (DGX Spark and partners)
- Memory
- 128 GB
- Bandwidth
- 273 GB/s
Largest popular model at 4-bit: Qwen3.5 122B A10B
128 GB with CUDA in a tiny box: the NVIDIA answer to Strix Halo and the Mac.
from $5,000 (typical price)Recommended