AM AMD Consumer Discrete GPU

AMD Radeon RX 7900 XTX

Runs downloadable LLMs up to ~32.5B at Q4, ~135-165 tok/s on an 8B model.

AMD Radeon RX 7900 XTX
24GB
VRAM
960
GB/s bandwidth
32.5B
Max model (Q4)
$999
MSRP

What you can run

At 8k context, FP16 KV. Full-speed tok/s holds up to "max ctx"; beyond it the context spills to system RAM (CPU/PCIe offload) and drops to the degraded figure.

Model sizeFP16Q8Q4
3B 100.8-123.2 t/s
≤222k ctx · then ~4.2-5.1 (slow)
201.6-246.4 t/s
≤268k ctx · then ~8.4-10.3 (slow)
360-440 t/s
≤288k ctx · then ~15-18.3 (slow)
7-8B 37.8-46.2 t/s
≤35k ctx · then ~1.6-1.9 (slow)
75.6-92.4 t/s
≤96k ctx · then ~3.2-3.9 (slow)
135-165 t/s
≤123k ctx · then ~5.6-6.9 (slow)
13B won't run 46.5-56.9 t/s
≤46k ctx · then ~1.9-2.4 (slow)
83.1-101.5 t/s
≤81k ctx · then ~3.5-4.2 (slow)
32B won't run won't run 33.8-41.3 t/s
≤10k ctx · then ~1.4-1.7 (slow)
70B won't run won't run won't run
120B won't run won't run won't run
400B won't run won't run won't run

Estimate: tok/s ≈ bandwidth ÷ (params × bytes/param) × efficiency (AMD). Cyan = fits ~22.1 GB usable. Beyond "ctx", KV spills to system RAM (CPU/PCIe offload) (rough degraded speed shown).

What else you need for a full build

Minimum supporting components for a desktop built around the AMD Radeon RX 7900 XTX. Guidance — exact needs depend on your case, other parts and how much you offload to system RAM.

Power supply
≥ 850W · 80+ Gold
Headroom over the 355W card plus the rest of the build.
System RAM
≥ 32 GB · 64 GB recommended
At least the 24 GB of VRAM; 2× lets you CPU-offload models larger than VRAM.
Motherboard
1× free PCIe x16 slot (4.0/5.0)
A full-length x16 electrical slot; PCIe 4.0 is ample, 5.0 future-proofs.
CPU
Any modern multi-core
The GPU runs inference; more cores and memory channels only matter when offloading to RAM.
Storage
≥ 1 TB NVMe SSD
Weights are large — a 70B model at Q4 is ~40 GB; budget for a few.

Run popular models on the AMD Radeon RX 7900 XTX

Videos about the AMD Radeon RX 7900 XTX

7900 XTX + Instinct MI50-32gb: AMD's Unholy LLM Alliance. ROCm ROCon.

AMD Radeon RX 7900 XTX Review & GPU Benchmarks: Gaming, Thermals, Power, & Noise

More hardware like the AMD Radeon RX 7900 XTX

About the AMD Radeon RX 7900 XTX

AMD Radeon RX 7900 XTX is a discrete GPU from AMD on RDNA 3, 2022. For local genAI the numbers that matter are its 24 GB (how big a model + context fits) and 960 GB/s bandwidth (generation speed). Plan for a ~850W power supply.

Frequently asked questions

What size LLM can the AMD Radeon RX 7900 XTX run?

With ~22.1 GB usable it runs up to ~32.5B at Q4 (~9.5B at FP16) at 8k context. Bigger spills into system RAM (CPU/PCIe offload) and slows sharply.

How many tokens/sec does the AMD Radeon RX 7900 XTX generate?

Generation is bandwidth-bound. At 960 GB/s an 8B model at Q4 runs ~135-165 tok/s (estimate).

Why does long context slow it down?

The KV cache for the context must also fit fast memory; once weights+KV exceed 22.1 GB it spills to system RAM (CPU/PCIe offload) and throughput collapses. Each row shows the max usable context.

What else do I need to run the AMD Radeon RX 7900 XTX?

Besides the card: a ≥850W 80+ Gold power supply, at least 24GB of system RAM (2× the 24GB VRAM is ideal for CPU offload), a free PCIe x16 slot, any modern multi-core CPU, and a ≥1TB NVMe SSD for model weights.