AMD Instinct MI350X 288GBBudget pick
C234.1 tok/s decode
~$8,000 MSRP
Can it run?
Runs with offload
Using Q4_K_M in vLLM
Fit status
Runs with offload
Decode
18.6 tok/s
TTFT
10410 ms
Safe context
16K
Memory
89.8 GB / 92.2 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs with offload | 18.6 tok/s | 15141 ms | 32K |
| Chat | C | Runs with offload | 18.6 tok/s | 5678 ms | 8K |
| Coding | C | Runs with offload | 18.6 tok/s | 10410 ms | 16K |
| RAG | C | Runs with offload | 18.6 tok/s | 18927 ms | 32K |
| Reasoning | C | Runs with offload | 18.6 tok/s | 12302 ms | 16K |
How Mistral Small 4 119B (119B params) fits at each quantization level on Mac Studio M2 Ultra 128GB (92.2 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 46.4 GB | Low | C40 |
Q3_K_S | 3 | 58.3 GB | Low | C43 |
NVFP4Best for your GPU | 4 | 66.6 GB | Medium | C45 |
Q4_K_M | 4 | 72.6 GB | Medium | C45 |
Q5_K_M | 5 | 85.7 GB | High | C45 |
Q6_K | 6 | 97.6 GB | High | F0 |
Q8_0 | 8 | 127.3 GB | Very High | F0 |
F16 | 16 | 244.0 GB | Maximum | F0 |
huggingface-cli download mistral-small-4-119bUpgrade options
~$8,000 MSRP
~$15,000 MSRP
Gaudi 3 128GBBiggest leap