AMD Radeon RX 5500 XT 8GB
No editorial image yet — generic vendor mark shown. Credentials in spec table below.
RDNA 1 entry. 8 GB GDDR6 at 224 GB/s. ROCm support was always experimental on RDNA 1 and is effectively defunct in 2026 — Vulkan via llama.cpp is the only operator-grade path. Performance ~12-22 tok/s on 7B Q4. Buy this card for AI? No. Have one already and want to tinker? Yes, with limits.
Affiliate disclosure: as an Amazon Associate and partner of other retailers, we earn from qualifying purchases. The verdict on this page is our editorial opinion; affiliate links never influence what we recommend.
Sub-scores sum to 239 / 1000. Headline = 239 × 0.70 (Estimated-confidence discount) = 167. This is an algorithmic performance-tier score — distinct from, and often lower than, the editorial “Our verdict” below, which weighs value and real-world fit (especially for hardware we haven’t measured yet). How scoring works →
Extrapolated from 224 GB/s bandwidth — 22.4 tok/s estimated. No measured benchmarks yet.
Plain-English: Edge-of-fit for 7B; expect compromises.
Verdicts extrapolated from catalog VRAM + bandwidth + ecosystem flags. Hover any chip for the rationale. Want measured numbers? Submit your own run with runlocalai-bench --submit.
This card is for the tinkerer who already owns one and wants to see what local AI looks like on a budget, not for anyone buying hardware specifically for inference. On a 7B Q4 model, expect 12-22 tok/s via Vulkan in llama.cpp—usable for chat but not real-time. The 8 GB VRAM can run 13B models at Q4 with tight context, but performance drops to 8-12 tok/s. ROCm support is effectively dead on RDNA 1, so Vulkan is the only path; no CUDA, no official ROCm. Pass if you're buying new—an RX 6600 or used RTX 3060 12GB offers far better performance and software support for a small premium. At ~$110 used, it's a curiosity, not a workhorse.
›Why this rating
The RX 5500 XT's limited VRAM and defunct ROCm support severely constrain its local AI utility. It's only viable for lightweight models via Vulkan, and even then performance is low. The rating reflects its status as a budget curiosity, not a practical inference card.
Overview
RDNA 1 entry. 8 GB GDDR6 at 224 GB/s. ROCm support was always experimental on RDNA 1 and is effectively defunct in 2026 — Vulkan via llama.cpp is the only operator-grade path. Performance ~12-22 tok/s on 7B Q4. Buy this card for AI? No. Have one already and want to tinker? Yes, with limits.
Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.
Specs
| VRAM | 8 GB |
| Power draw (peak) | 130 W |
| Released | 2019 |
| MSRP | $199 |
| Backends | Vulkan |
Models that fit
Open-weight models small enough to run on AMD Radeon RX 5500 XT 8GB with usable context.
Frequently asked
What models can AMD Radeon RX 5500 XT 8GB run?
Does AMD Radeon RX 5500 XT 8GB support CUDA?
How much does AMD Radeon RX 5500 XT 8GB cost?
Where next?
Reviewed by RunLocalAI Editorial. See our editorial policy for how we research and verify hardware specifications.