RX 7900 XTX 24GBBest value
C41.5 tok/s decode
~$999 MSRP
Can it run?
Too heavy
Using Q4_K_M in Ollama
Fit status
Too heavy
Decode
24.9 tok/s
TTFT
7785 ms
Safe context
11K
Memory
23.5 GB / 16.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | F | Too heavy | 24.9 tok/s | 11323 ms | 18K |
| Chat | F | Too heavy | 24.9 tok/s | 4246 ms | 6K |
| Coding | F | Too heavy | 24.9 tok/s | 7785 ms | 11K |
| RAG | F | Too heavy | 24.9 tok/s | 14154 ms | 18K |
| Reasoning | F | Too heavy | 24.9 tok/s | 9200 ms | 11K |
How gemma 3 27b it (27B params) fits at each quantization level on RX 9070 XT 16GB (16.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_KBest for your GPU | 2 | 10.5 GB | Low | C43 |
Q3_K_S | 3 | 13.2 GB | Low | C44 |
NVFP4 | 4 | 15.1 GB | Medium | C44 |
Q4_K_M | 4 | 16.5 GB | Medium | F0 |
Q5_K_M | 5 | 19.4 GB | High | F0 |
Q6_K | 6 | 22.1 GB | High | F0 |
Q8_0 | 8 | 28.9 GB | Very High | F0 |
F16 | 16 | 55.4 GB | Maximum | F0 |
Upgrade options
~$999 MSRP
~$1,099 MSRP
RTX 5090 32GBBiggest leap~$1,999 MSRP