Hardware › AMD Radeon RX 7900 XTX 24GB
AMD Radeon RX 7900 XTX 24GB for local AI: price in Australia & what it can run
price unavailableno AU retailer price found
24 GBVRAM (23.5 GB usable)
960 GB/smemory bandwidth
183 tok/smeasured, Llama 2 7B Q4_0
Discontinued in most AU stores.
Benchmark
182.63 tok/s on Llama 2 7B Q4_0, tg128 (llama.cpp llama-bench), Vulkan backend. Source: ggml-org/llama.cpp/discussions/10879 scoreboard (checked 2026-10-19). ROCm scoreboard: 167.11 tok/s.
Specs: AMD published specification (manually entered)
AU prices found
Price unavailable: no matching listing at the retailers we can check.
Community measured
No reader results yet. Be the first to submit one →
Which models fit? (8k context)
| Model | Quant | Needs / usable | Fit | Est. tok/s |
|---|---|---|---|---|
Q4_K_M | 464.4 / 23.5 GB | no | – | |
Q4_K_M | 499.4 / 23.5 GB | no | – | |
Q4_K_M | 20.0 / 23.5 GB | fits | 381 estimate | |
Q4_K_M | 348.5 / 23.5 GB | no | – | |
Q4_K_M | 15.7 / 23.5 GB | fits | 571 estimate | |
Q4_K_M | 45.9 / 23.5 GB | no | – | |
Q4_K_M | 22.5 / 23.5 GB | tight | 381 estimate | |
Q4_K_M | 305.3 / 23.5 GB | no | – | |
MiniCPM V 4.7 1B new | Q4_K_M | 2.0 / 23.5 GB | fits | 737 estimate |
Q4_K_M | 464.4 / 23.5 GB | no | – | |
Q4_K_M | 195.9 / 23.5 GB | no | – | |
GLM 5.3 (MoE) new | Q4_K_M | 458.4 / 23.5 GB | no | – |
Q4_K_M | 110.3 / 23.5 GB | no | – | |
Q4_K_M | 18.4 / 23.5 GB | fits | 41 estimate | |
Q4_K_M | 20.2 / 23.5 GB | fits | 381 estimate | |
Q4_K_M | 9.1 / 23.5 GB | fits | 95 estimate | |
Q4_K_M | 23.0 / 23.5 GB | tight | 381 estimate | |
Q4_K_M | 81.4 / 23.5 GB | no | – | |
Q4_K_M | 22.1 / 23.5 GB | tight | 37 estimate | |
Q4_K_M | 17.2 / 23.5 GB | fits | 301 estimate | |
Q4_K_M | 6.0 / 23.5 GB | fits | 143 estimate | |
Q4_K_M | 7.1 / 23.5 GB | fits | 118 estimate | |
Q4_K_M | 4.1 / 23.5 GB | fits | 245 estimate | |
Q4_K_M | 77.0 / 23.5 GB | no | – | |
Q4_K_M | 20.4 / 23.5 GB | fits | 381 estimate | |
gpt-oss 20B (MoE) older | MXFP4 | 12.4 / 23.5 GB | fits | 362 estimate |
gpt-oss 120B (MoE) older | MXFP4 | 63.5 / 23.5 GB | no | – |
Qwen 3 4B older | Q4_K_M | 4.6 / 23.5 GB | fits | 284 estimate |
Qwen 3 8B older | Q4_K_M | 7.2 / 23.5 GB | fits | 139 estimate |
Qwen 3 32B older | Q4_K_M | 23.0 / 23.5 GB | tight | 35 estimate |
Llama 3.1 8B Instruct older | Q4_K_M | 6.9 / 23.5 GB | fits | 142 estimate |
Mistral 7B Instruct v0.3 older | Q4_K_M | 6.5 / 23.5 GB | fits | 158 estimate |