What AI models can the RX 7900 XT run?
The Radeon RX 7900 XT has 20 GB of GDDR6 at 800 GB/s. With an 8K-token context it fits 15 of 21 models we track at Q4, 11 at Q8 and 6 at FP16; the largest text model it runs at Q4 is Qwen3 30B-A3B (MoE). The cheapest price we found today is $770 (used, GPU Sold, 2026-11-10).
Prices updated Nov 12, 2026 · ROCm/Vulkan - works in llama.cpp & Ollama; image/video tooling less matureEvery model we track on the RX 7900 XT
| Model | Q4 | Q8 | FP16 |
|---|---|---|---|
| Mistral 7B Instruct v0.37.2B | ✓ ~87 tok/s6.1 GB | ✓ ~54 tok/s9.2 GB | ✓ ~30 tok/s15.5 GB |
| Llama 3.1 8B Instruct8.0B | ✓ ~80 tok/s6.5 GB | ✓ ~49 tok/s9.9 GB | ✓ ~28 tok/s17.0 GB |
| DeepSeek R1 Distill Llama 8B8.0B | ✓ ~80 tok/s6.5 GB | ✓ ~49 tok/s9.9 GB | ✓ ~28 tok/s17.0 GB |
| Qwen3 8B8.2B | ✓ ~77 tok/s6.7 GB | ✓ ~48 tok/s10.2 GB | ✓ ~27 tok/s17.4 GB |
| Gemma 3 12B IT12.2B | ✓ ~58 tok/s8.7 GB | ✓ ~34 tok/s13.9 GB | ✗needs 25 GB |
| Qwen3 14B14.8B | ✓ ~46 tok/s10.6 GB | ✓ ~28 tok/s16.9 GB | ✗needs 30 GB |
| DeepSeek R1 Distill Qwen 14B14.8B | ✓ ~45 tok/s10.8 GB | ✓ ~27 tok/s17.1 GB | ✗needs 30 GB |
| Mistral Small 24B Instruct (2501)23.6B | ✓ ~30 tok/s15.6 GB | ✗needs 26 GB | ✗needs 46 GB |
| Gemma 3 27B IT27.4B | ✓ ~27 tok/s17.5 GB | ✗needs 29 GB | ✗needs 53 GB |
| Qwen3 30B-A3B (MoE)30.5B | ✓ ~169 tok/s19.0 GB | ✗needs 32 GB | ✗needs 59 GB |
| Qwen3 32B32.8B | ✗needs 21 GB | ✗needs 35 GB | ✗needs 64 GB |
| Qwen2.5 Coder 32B Instruct32.8B | ✗needs 21 GB | ✗needs 35 GB | ✗needs 64 GB |
| DeepSeek R1 Distill Qwen 32B32.8B | ✗needs 21 GB | ✗needs 35 GB | ✗needs 64 GB |
| Mixtral 8x7B Instruct (MoE)46.7B | ✗needs 28 GB | ✗needs 48 GB | ✗needs 89 GB |
| Llama 3.3 70B Instruct70.6B | ✗needs 43 GB | ✗needs 73 GB | ✗needs 135 GB |
| Qwen3 235B-A22B (MoE)235.1B | ✗needs 135 GB | ✗needs 235 GB | ✗needs 440 GB |
| FLUX.1 [dev]11.9B · image | ✓ fits9.9 GB | ✓ fits15.0 GB | ✗needs 25 GB |
| FLUX.1 [schnell]11.9B · image | ✓ fits9.9 GB | ✓ fits15.0 GB | ✗needs 25 GB |
| Stable Diffusion XL 1.02.6B · image | ✓ fits5.2 GB | ✓ fits6.3 GB | ✓ fits8.6 GB |
| Wan 2.1 T2V 14B (video)14.3B · video | ✓ fits14.9 GB | ✗needs 21 GB | ✗needs 33 GB |
| Wan 2.1 T2V 1.3B (video)1.4B · video | ✓ fits4.6 GB | ✓ fits5.2 GB | ✓ fits6.4 GB |
✓ = fits in 20 GB with an 8K-token context (text models) and ~1 GB runtime overhead; tok/s is a rough single-user estimate (0.6 × bandwidth ÷ bytes per token), not a benchmark. Leave ~0.5–1 GB free if this card also drives your display. Method.
Similar-VRAM cards, cheapest first
FAQ
How much VRAM does the RX 7900 XT have?
20 GB of GDDR6 with 800 GB/s of memory bandwidth on a 320-bit bus.
What is the biggest AI model the RX 7900 XT can run?
Of the models we track, Qwen3 30B-A3B (MoE) (31B) is the largest text model that fits at 4-bit (Q4) with an 8K context. Bigger models need a second GPU or CPU offload.
Can the RX 7900 XT run Llama 3.3 70B Instruct?
Not on a single card: it needs about 43 GB at Q4 and the RX 7900 XT has 20 GB.
Can the RX 7900 XT run Qwen3 32B?
Not on a single card: it needs about 21 GB at Q4 and the RX 7900 XT has 20 GB.
Can the RX 7900 XT run FLUX.1 [dev]?
Yes, at Q4, Q8 (about 10 GB needed at Q4).
How much does the RX 7900 XT cost right now?
About $770 (used, GPU Sold, observed 2026-11-10). Prices change daily; check the retailer.
Best GPU for…
Get the weekly GPU price drops
Cheapest card per model, new and used. Free, 1 email a week.
Subscribe on beehiiv →