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-nano-omni-30b-a3b-reasoning

Serving

Rank 5 of 16 Last probed (22 min ago)

nvidia/nemotron-3-nano-omni-30b-a3b-reasoning answered 81.1% of probes on the NVIDIA NIM free API in the last 30 days, with a median time to first token of 814 ms and 81.8 tok/s. It ranks 5 of 16 tracked models.

81.1% Uptime 575/709 probes
78/ 100 Score 342 round wins
814 ms First token p95 3.61 s
6.14 s Full reply median
1.51 s Best reply fastest in range
81.8tok/s Throughput median

Model vs. median

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

Uptime
81.1% 77.3%
4 pts above median
First token
814 ms 16.6 s
20× better than median
Full reply
6.14 s 22.6 s
3.7× better than median
Throughput
81.8 tok/s 18.0 tok/s
4.5× better than median

nemotron-3-nano-omni-30b-a3b-reasoning median of answering models

Error breakdown

134 failed of 709 probes, last 30 days.

FailureProbesShare of failures
Empty answer132 98.5%
Server error (5xx)1 0.7%
Timed out1 0.7%

Time to first token

Each probe. Lower is better.

0 ms25.0 s50.0 s75.0 s100.0 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
OK180 ms2.05 s107 tok/s
OK176 ms1.77 s126 tok/s
OK215 ms2.85 s75.8 tok/s
OK179 ms2.55 s84.2 tok/s
OK193 ms2.85 s75.2 tok/s
OK148 ms1.84 s118 tok/s
OK170 ms2.08 s105 tok/s
OK183 ms2.39 s90.5 tok/s
OK280 ms3.04 s72.6 tok/s
OK241 ms4.31 s49.2 tok/s
OK145 ms2.11 s102 tok/s
OK147 ms2.00 s108 tok/s
OK972 ms9.24 s54.1 tok/s
OK1.44 s5.43 s92.1 tok/s
OK192 ms2.03 s109 tok/s
OK710 ms6.20 s80.7 tok/s
OK665 ms5.42 s92.3 tok/s
OK710 ms4.87 s103 tok/s
OK407 ms4.09 s122 tok/s
OK621 ms3.99 s125 tok/s

Call nemotron-3-nano-omni-30b-a3b-reasoning

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-nano-omni-30b-a3b-reasoning",
    "messages": [{"role": "user", "content": "Hello"}],
    "max_tokens": 256,
    "stream": true
  }'

Other models answering

Top of the leaderboard right now.