gpt-oss 120BNVIDIA Blackwell
Running gpt-oss-120b on GB200 NVL72
Quick answer
gpt-oss-120b runs on GB200 NVL72: 34 benchmarked configs so far. See the interactivity ladder below for measured operating points.
Benchmarked configs
34
Serving engines
Dynamo TRTLLM
Precisions
fp4
Run dates
2026-01-07 → 2026-01-07
Throughput at every interactivity target
Serving is a trade-off: push more concurrent users through a GPU and each user's tokens arrive slower. The ladder below reads the measured frontier at each per-user speed target on a single-turn chat workload (8k input / 1k output), using the best engine and precision at that point.
| Per-user target | Tokens/s per GPU | $ / 1M tokens | Engine | Precision |
|---|---|---|---|---|
| 30 tok/s | - | - | - | - |
| 50 tok/s | - | - | - | - |
| 75 tok/s | - | - | - | - |
| 100 tok/s | - | - | - | - |
| 150 tok/s | - | - | - | - |
| 200 tok/s | - | - | - | - |
Frequently asked questions
- How fast is gpt-oss-120b on GB200 NVL72?
- The InferenceX fleet has 34 benchmarked configs for this pairing; see the interactivity ladder above for the operating points reached so far.
- How much does it cost to serve gpt-oss-120b on GB200 NVL72?
- Cost per million tokens is derived from measured throughput and $/GPU/hr rates from the SemiAnalysis AI Cloud TCO model; it appears once this pairing reaches the primary interactivity tier.
- Which serving engines run gpt-oss-120b on GB200 NVL72?
- The runs behind this page used Dynamo TRTLLM in FP4, including disaggregated prefill and multi-node serving. Engines are rebuilt and re-benchmarked continuously, so the best config can change between visits.
- How are these gpt-oss-120b numbers measured?
- Every number is measured on real GB200 NVL72 hardware by the InferenceX fleet, sweeping concurrency on a single-turn chat workload (8k input / 1k output) to trace the throughput-versus-interactivity frontier; the newest run landed on 2026-01-07. The same derivation powers the InferenceX overview leaderboard.