RX 7900 XTX 24GBBudget pick
B56.7 tok/s decode
~$999 MSRP
Can it run?
Tight fit
Using Q4_K_M in Ollama
Fit status
Tight fit
Decode
7.1 tok/s
TTFT
27337 ms
Safe context
8K
Memory
20.0 GB / 23.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs with offload (needs ~0 GB host RAM) | 7.1 tok/s | 39797 ms | 8K |
| Chat | C | Runs well | 7.1 tok/s | 14911 ms | 8K |
| Coding | C | Tight fit | 7.1 tok/s | 27337 ms | 8K |
| RAG | C | Runs with offload (needs ~0 GB host RAM) | 7.1 tok/s | 49746 ms | 8K |
| Reasoning | C | Tight fit | 7.1 tok/s | 32307 ms | 8K |
How Granite Code 20B (20B params) fits at each quantization level on MacBook Pro M4 32GB (23.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 7.8 GB | Low | D36 |
Q3_K_S | 3 | 9.8 GB | Low | D38 |
NVFP4 | 4 | 11.2 GB | Medium | D39 |
Q4_K_M | 4 | 12.2 GB | Medium | C40 |
Q5_K_M | 5 | 14.4 GB | High | C42 |
Q6_KBest for your GPU | 6 | 16.4 GB | High | C44 |
Q8_0 | 8 | 21.4 GB | Very High | C44 |
F16 | 16 | 41.0 GB | Maximum | F0 |
ollama run granite-code-20bhuggingface-cli download granite-code-20bUpgrade options
~$999 MSRP
RTX 3090 24GBBest value~$1,499 MSRP
RTX 4090 24GBBiggest leap~$1,599 MSRP
~$2,499 MSRP