AMD GPUs
AMD Radeon RX 7900 XTX
- Memory
- 24 GB
- Bandwidth
- 960 GB/s
Largest popular model at 4-bit: Qwen3.6 27B
24 GB and 960 GB/s: RTX 3090-class memory, bought new.
from $1,358 (typical price)Recommended
NVIDIA · NVIDIA GPUs · 2025
NVIDIA GeForce RTX 5070 Ti
Illustration
Recommended
Nearly RTX 5080 bandwidth for less. Still 16 GB, so speed rather than size.
- Memory
- 16 GB
- Bandwidth
- 896 GB/s
- Power
- 300 W
- Largest popular model at 4-bit
- gpt-oss-20b
from $1,190 (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 | ≤ 807 tok/s | |
| Qwen3.5 9B | Runs | ≤ 158 tok/s | |
| Gemma 4 12B | Runs | ≤ 126 tok/s | |
| gpt-oss-20b | Runs | ≤ 448 tok/s | |
| Qwen3.6 27B | Not enough memory | — | |
| Qwen3.6 35B A3B | Partly in system RAM | — | |
| 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 5070 Ti 16 GB (any partner card)16 GBLaunch price: $749 · Typical price: ~$1,190 (checked Oct 9, 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: Fast answers from 8–14B models and MoE models that fit in 16 GB.
Editorial opinionWould we buy it today?
A good fast card for small models; for more memory per euro look at 24–32 GB options.
Pros and cons
Editorial opinionPros
- 896 GB/s: quick token generation
- Same memory as the RTX 5080 for less money
Cons
- 16 GB is the ceiling
- Prices well above launch in 2026
Specifications
Factual data- Manufacturer
- NVIDIA
- Chip
- GB203
- Architecture
- Blackwell
- Type
- Graphics card
- Form factor
- PCIe card
- Released
- 2025
- Availability
- On sale
- Memory options
- 16 GB
- Memory type
- GDDR7
- Memory bus
- 256-bit
- Memory bandwidth
- 896 GB/s
- Usable for models (default)
- 16 GB / 16 GB
- Board / max power
- 300 W
- Interface
- PCIe 5.0 x16
- 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
AMD GPUs
AMD Radeon RX 9070 XT
- Memory
- 16 GB
- Bandwidth
- 645 GB/s
Largest popular model at 4-bit: gpt-oss-20b
AMD's best gaming card is a solid 16 GB AI card too, with ROCm and Vulkan.
from $759 (typical price)Recommended
NVIDIA GPUs
NVIDIA RTX PRO 4000 Blackwell
- Memory
- 24 GB
- Bandwidth
- 672 GB/s
Largest popular model at 4-bit: Qwen3.6 27B
24 GB in a single-slot, 140 W card: the homelab-friendly way to get 24 GB new.
from $1,500 (typical price)Recommended