Can it run?
Runs well
Using Q3_K_M in Ollama
Fit status
Runs well
Decode
89.6 tok/s
TTFT
2952 ms
Safe context
31K
Memory
99.2 GB / 192.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | B | Runs well | 104.5 tok/s | 2694 ms | 52K |
| Chat | C | Runs well | 104.5 tok/s | 1010 ms | 17K |
| Coding | C | Runs well | 89.6 tok/s | 2952 ms | 31K |
| RAG | B | Runs well | 104.5 tok/s | 3367 ms | 52K |
| Reasoning | C | Runs well | 104.5 tok/s | 2189 ms | 31K |
How Qwen3.5 122B A10B (122B params) fits at each quantization level on NVIDIA GB200 192GB (192.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 47.6 GB | Low | D35 |
Q3_K_S | 3 | 59.8 GB | Low | D36 |
NVFP4 | 4 |
huggingface-cli download hf-unsloth--qwen3-5-122b-a10b-gguf68.3 GB |
| Medium |
| D37 |
Q4_K_M | 4 | 74.4 GB | Medium | D38 |
Q5_K_M | 5 | 87.8 GB | High | D39 |
Q6_K | 6 | 100.0 GB | High | C41 |
Q8_0Best for your GPU | 8 | 130.5 GB | Very High | C44 |
F16 | 16 | 250.1 GB | Maximum | F0 |