AMD Radeon RX 7900 XTX
24 GB · 960 GB/s · 355 W
Prices
Every retailer we track in the US, cheapest first. Buy links may earn us a commission.
No retailer in this market lists it right now.
What it runs
Every model in the catalogue, with the best quant for this hardware. Estimates, not measurements.
| Model | Quant | Fits | Memory | Max context | Decode @8k | Prefill | Good for |
|---|---|---|---|---|---|---|---|
| Qwen 3.8 27B Alibaba Qwen | UD-Q6_K | fits | 22.0 GB | 26k45k with q8 cache | 23tok/sest. | 753 tok/s | chat |
| Qwen 3.6 27B Alibaba Qwen | Q6_K | fits | 22.5 GB | 18k32k with q8 cache | 23tok/sest. | 753 tok/s | chat |
| Gemma 4 31B | Q5_K_M | fits | 22.4 GB | 17k38k with q8 cache | 23tok/sest. | 657 tok/s | chat |
| Qwen 3.6 35B-A3B Alibaba Qwen | UD-Q4_K_M | fits | 21.7 GB | 68k106k with q8 cache | 153tok/sest. | 6453 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 27B Alibaba Qwen | Q6_K | fits | 22.4 GB | 19k34k with q8 cache | 23tok/sest. | 753 tok/s | chat |
| Gemma 4 26B-A4B | UD-Q6_K | fits | 22.8 GB | 24k44k with q8 cache | 80tok/sest. | 4917 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 35B-A3B Alibaba Qwen | Q4_K_M | fits | 21.6 GB | 72k113k with q8 cache | 153tok/sest. | 6453 tok/s | chat, coding, agents, long documents |
| GLM 4.7 Flash Z.ai | Q5_K_M | fits | 21.2 GB | 42k73k with q8 cache | 79tok/sest. | 3611 tok/s | chat, coding, agents |
| Devstral Small 2 24B Mistral AI | Q6_K | fits | 20.1 GB | 27k49k with q8 cache | 26tok/sest. | 849 tok/s | chat |
| gpt-oss 20B OpenAI | MXFP4 | fits | 12.3 GB | 128k | 120tok/sest. | 4671 tok/s | chat, coding, agents, long documents |
| Qwen3 Coder 30B-A3B Alibaba Qwen | Q5_K_M | fits | 21.8 GB | 21k38k with q8 cache | 108tok/sest. | 4669 tok/s | chat, coding, agents |
| Qwen3 32B Alibaba Qwen | Q4_K_M | fits | 21.3 GB | 15k28k with q8 cache | 24tok/sest. | 581 tok/s | chat |
| Gemma 4 12B | Q8_0 | fits | 13.1 GB | 256k | 40tok/sest. | 1713 tok/s | chat, coding, long documents |
| Qwen 3.5 9B Alibaba Qwen | Q8_0 | fits | 10.0 GB | 256k | 54tok/sest. | 2170 tok/s | chat, coding, long documents |
| Mistral Small 3.2 24B Mistral AI | Q6_K | fits | 20.1 GB | 27k49k with q8 cache | 26tok/sest. | 849 tok/s | chat |
| DeepSeek R1 Distill Qwen 32B DeepSeek | Q4_K_M | fits | 21.3 GB | 15k28k with q8 cache | 24tok/sest. | 607 tok/s | chat |
| Gemma 3 27B | Q6_K | fits | 22.5 GB | 16k32k with q8 cache | 23tok/sest. | 763 tok/s | chat |
| Ministral 3 14B Mistral AI | Q8_0 | fits | 15.5 GB | 55k101k with q8 cache | 34tok/sest. | 1408 tok/s | chat, coding |
| Qwen3 14B Alibaba Qwen | Q8_0 | fits | 16.7 GB | 40k | 31tok/sest. | 1309 tok/s | chat, coding |
| DeepSeek R1 0528 Qwen3 8B DeepSeek | Q8_0 | fits | 10.1 GB | 97k128k with q8 cache | 53tok/sest. | 2290 tok/s | chat, coding |
| Qwen3 8B Alibaba Qwen | Q8_0 | fits | 10.1 GB | 40k | 53tok/sest. | 2290 tok/s | chat, coding |
| Phi-4 14B Microsoft | Q8_0 | fits | 16.9 GB | 16k | 31tok/sest. | 1318 tok/s | chat, coding |
| Qwen 3.5 4B Alibaba Qwen | Q8_0 | fits | 5.3 GB | 256k | 110tok/sest. | 4368 tok/s | chat, coding, long documents |
| Ministral 3 8B Mistral AI | Q8_0 | fits | 10.3 GB | 100k183k with q8 cache | 52tok/sest. | 2140 tok/s | chat, coding |
| Granite 4.2 8B IBM | Q8_0 | fits | 10.8 GB | 84k128k with q8 cache | 49tok/sest. | 2124 tok/s | chat, coding, agents |
| Gemma 4 E4B | Q8_0 | fits | 8.7 GB | 128k | 62tok/sest. | 2629 tok/s | chat, coding, long documents |
| LFM2.5 8B-A1B Liquid AI | Q8_0 | fits | 9.6 GB | 125k | 216tok/sest. | 15346 tok/s | chat, coding, long documents |
| Llama 3.1 8B Meta | Q6_K | fits | 8.0 GB | 124k128k with q8 cache | 69tok/sest. | 2364 tok/s | chat, coding, long documents |
| Qwen 3.5 122B-A10B Alibaba Qwen | UD-Q2_K_XL | offload | 40.1 GB | – | 18tok/sest. | 1477 tok/s | chat |
| Qwen3 Coder Next Alibaba Qwen | Q4_K_M | offload | 46.3 GB | – | 22tok/sest. | 3498 tok/s | chat |
| Mistral Small 4 119B Mistral AI | UD-Q2_K_XL | offload | 38.5 GB | – | 27tok/sest. | 1969 tok/s | chat, coding |
| GLM 4.5 Air Z.ai | Q2_K | offload | 44.5 GB | – | 7tok/sest. | 703 tok/s | – |
| Qwen3 Next 80B-A3B Alibaba Qwen | Q4_K_M | offload | 46.3 GB | – | 28tok/sest. | 4398 tok/s | chat, coding |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA | UD-Q6_K_XL | offload | 33.5 GB | – | 24tok/sest. | 5448 tok/s | chat, coding, long documents |
| Kimi Linear 48B-A3B Moonshot AI | Q4_K_M | offload | 28.9 GB | – | 58tok/sest. | 5381 tok/s | chat, coding, long documents |
| Llama 3.3 70B Meta | Q4_K_M | offload | 42.9 GB | – | 2tok/sest. | 204 tok/s | – |
| DeepSeek V4 Flash DeepSeek | UD-Q2_K_XL | won't run | 91.7 GB | – | – | – | – |
| GLM 5.3 Flash Z.ai | UD-Q2_K_XL | won't run | 102.3 GB | – | – | – | – |
| Qwen 3.8 Flash Next Alibaba Qwen | UD-Q2_K_XL | won't run | 74.6 GB | – | – | – | – |
| MiniMax M2.7 MiniMax | UD-Q2_K_XL | won't run | 72.9 GB | – | – | – | – |
| Nemotron 3 Super 120B-A12B NVIDIA | UD-Q2_K_XL | won't run | 52.0 GB | – | – | – | – |
| gpt-oss 120B OpenAI | MXFP4 | won't run | 60.2 GB | – | – | – | – |
Specifications
From the manufacturer's published specifications; sources are linked where we have them.
- GPU
- Radeon RX 7900 XTX
- Memory
- 24 GB GDDR6
- Memory bus
- 384-bit
- Memory bandwidth
- 960 GB/s
- FP16 compute
- 123 TFLOPS
- Software
- ROCm
- Board power
- 355 W
- PCIe generation
- Gen 4
- PCIe lanes
- x16