GPT-OSS 120B needs ~95.2 GB VRAM. NVIDIA B200 180GB has 180.0 GB. With Q4_K_M quantization, expect ~102 tok/s.
Operating mode
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
Fit status
Runs well
Decode
102.4 tok/s
TTFT
1891 ms
Safe context
131K
Memory
95.2 GB / 180.0 GB
This setup is broadly balanced for this model.
No major red flags
This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | S | Runs well | 102.4 tok/s | 1031 ms | 131K |
| Coding | S | Runs well | 102.4 tok/s | 1891 ms | 131K |
| Agentic Coding | S | Runs well | 102.4 tok/s | 2750 ms | 131K |
| Reasoning | S | Runs well | 102.4 tok/s | 2234 ms | 131K |
| RAG | S | Runs well | 102.4 tok/s | 3438 ms | 131K |
How GPT-OSS 120B (117B params) fits at each quantization level on NVIDIA B200 180GB (180.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 45.6 GB | Low | A82 |
Q3_K_S | 3 | 57.3 GB | Low | A83 |
NVFP4 | 4 |
Copy-paste commands to run GPT-OSS 120B on your machine.
Run
ollama run gpt-oss:120bYour hardware
| Model | Params | Grade | Decode | Capabilities |
|---|---|---|---|---|
| 123B | S | 97.4 tok/s | ||
| 122B | S |
Yes, NVIDIA B200 180GB can run GPT-OSS 120B with a S grade (Runs well). Expected decode speed: 102.4 tok/s.
GPT-OSS 120B (117B parameters) requires approximately 95.2 GB of memory with Q4_K_M quantization.
The recommended quantization for GPT-OSS 120B is Q4_K_M, which balances quality and memory efficiency.
On NVIDIA B200 180GB, GPT-OSS 120B achieves approximately 102.4 tokens per second decode speed with a time-to-first-token of 1891ms using Q4_K_M quantization.
For coding workloads, GPT-OSS 120B on NVIDIA B200 180GB receives a S grade with 102.4 tok/s and 131K context.
On NVIDIA B200 180GB, GPT-OSS 120B can safely use up to 131K tokens of context. The model's official context limit is 131K, but available memory constrains the safe maximum.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/gpt-oss-120b-on-b200-180gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview:
65.5 GB |
| Medium |
| A84 |
Q4_K_M | 4 | 71.4 GB | Medium | A85 |
Q5_K_M | 5 | 84.2 GB | High | S86 |
Q6_K | 6 | 95.9 GB | High | S88 |
Q8_0Best for your GPU | 8 | 125.2 GB | Very High | S88 |
F16 | 16 | 239.8 GB | Maximum | F0 |
| 270.2 tok/s |
| 284B | S | 144.8 tok/s |
| 119B | S | 292.9 tok/s |