Skip to main content

Includes 694 hourly rounds up to 2026-11-16 from NIMStats by Mauro Druwel (source), measured with a different prompt and timeout. Our own probe adds a round every 30 minutes.

nvidia

nemotron-3-ultra-550b-a55b

Serving

Rank 4 of 16 Last probed (22 min ago)

nvidia/nemotron-3-ultra-550b-a55b answered 94.9% of probes on the NVIDIA NIM free API in the last 30 days, with a median time to first token of 1.73 s and 48.5 tok/s. It ranks 4 of 16 tracked models.

94.9% Uptime 673/709 probes
82/ 100 Score 190 round wins
1.73 s First token p95 11.2 s
10.2 s Full reply median
1.79 s Best reply fastest in range
48.5tok/s Throughput median

Model vs. median

Against the median of models that answered, last 30 days.

Uptime
94.9% 77.3%
18 pts above median
First token
1.73 s 16.6 s
9.6× better than median
Full reply
10.2 s 22.6 s
2.2× better than median
Throughput
48.5 tok/s 18.0 tok/s
2.7× better than median

nemotron-3-ultra-550b-a55b median of answering models

Error breakdown

36 failed of 709 probes, last 30 days.

FailureProbesShare of failures
Empty answer19 52.8%
Server error (5xx)7 19.4%
Timed out6 16.7%
Network error4 11.1%

Time to first token

Each probe. Lower is better.

0 ms50.0 s100.0 s150.0 s200.0 s10-31 14:0011-17 16:13 UTC
answered no answer

Throughput

Tokens per second after the first token, each probe. Higher is better.

0.0 tok/s125 tok/s250 tok/s375 tok/s500 tok/s10-31 14:0011-17 16:13 UTC
answered no answer

Availability heatmap

Last 7 days by UTC hour.

all answered some failed none answered not probed

Run history

Last 20 probes.

TimeResultFirst tokenFull replySpeed
OK782 ms4.79 s50.0 tok/s
OK546 ms6.45 s33.0 tok/s
OK382 ms4.34 s50.6 tok/s
OK405 ms5.39 s40.1 tok/s
OK877 ms6.50 s35.6 tok/s
OK406 ms4.13 s53.6 tok/s
OK1.59 s4.87 s60.9 tok/s
OK397 ms2.50 s95.1 tok/s
OK461 ms6.72 s32.0 tok/s
OK414 ms7.04 s30.2 tok/s
OK369 ms13.7 s15.0 tok/s
OK717 ms3.83 s64.3 tok/s
OK2.64 s10.7 s46.9 tok/s
OK1.76 s10.7 s46.6 tok/s
OK436 ms3.64 s57.7 tok/s
OK3.56 s11.2 s44.7 tok/s
OK1.39 s15.6 s32.0 tok/s
OK5.59 s11.7 s42.9 tok/s
OK1.08 s8.68 s57.6 tok/s
OK2.51 s7.81 s64.0 tok/s

Call nemotron-3-ultra-550b-a55b

The endpoint is OpenAI-compatible.

Create a free key at build.nvidia.com, then:

curl https://integrate.api.nvidia.com/v1/chat/completions \
  -H "Authorization: Bearer $NVIDIA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/nemotron-3-ultra-550b-a55b",
    "messages": [{"role": "user", "content": "Hello"}],
    "max_tokens": 256,
    "stream": true
  }'

Other models answering

Top of the leaderboard right now.