MacBook Pro M3 Max 128GBBudget pick
C252 tok/s decode
~$2,499 MSRP
Can it run?
Runs well
Using Q6_K in Ollama
Fit status
Runs well
Decode
2954.7 tok/s
TTFT
350 ms
Safe context
125K
Memory
10.2 GB / 80.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 2532.6 tok/s | 350 ms | 250K |
| Chat | C | Runs well | 2954.7 tok/s | 350 ms | 62K |
| Coding | C | Runs well | 2954.7 tok/s | 350 ms | 125K |
| RAG | C | Runs well | 2954.7 tok/s | 350 ms | 250K |
| Reasoning | C | Runs well | 2954.7 tok/s | 350 ms | 125K |
How embeddinggemma 300M (0.30000001192092896B params) fits at each quantization level on NVIDIA H100 80GB (80.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 0.1 GB | Low | D29 |
Q3_K_S | 3 | 0.1 GB | Low | D29 |
NVFP4 | 4 |
huggingface-cli download hf-ggml-org--embeddinggemma-300m-ggufUpgrade options
~$2,499 MSRP
~$3,999 MSRP
| Medium |
| D29 |
Q4_K_M | 4 | 0.2 GB | Medium | D29 |
Q5_K_M | 5 | 0.2 GB | High | D29 |
Q6_K | 6 | 0.2 GB | High | D29 |
Q8_0 | 8 | 0.3 GB | Very High | D29 |
F16Best for your GPU | 16 | 0.6 GB | Maximum | D30 |
~$3,999 MSRP