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.

openai

gpt-oss-20b

Unstable

Rank 10 of 15 Last probed (22 min ago)

openai/gpt-oss-20b answered 77.2% of probes on the NVIDIA NIM free API in the last 14 days, with a median time to first token of 142.6 s and 3.4 tok/s. It ranks 10 of 15 tracked models.

77.2% Uptime 251/325 probes
51/ 100 Score 2 round wins
142.6 s First token p95 244.1 s
147.8 s Full reply median
6.98 s Best reply fastest in range
3.4tok/s Throughput median

Model vs. median

Against the median of models that answered, last 14 days.

Uptime
77.2% 79.4%
2 pts below median
First token
142.6 s 13.0 s
11× worse than median
Full reply
147.8 s 22.6 s
6.5× worse than median
Throughput
3.4 tok/s 21.3 tok/s
6.3× worse than median

gpt-oss-20b median of answering models

Error breakdown

74 failed of 325 probes, last 14 days.

FailureProbesShare of failures
Timed out64 86.5%
Network error6 8.1%
Empty answer4 5.4%

Time to first token

Each probe. Lower is better.

0 ms125.0 s250.0 s375.0 s500.0 s11-03 17:0011-17 16:13 UTC
answered no answer

Throughput

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

0.0 tok/s25.0 tok/s50.0 tok/s75.0 tok/s100 tok/s11-03 17:0011-17 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
Empty answern/an/an/a
Empty answern/an/an/a
Timed outn/an/an/a
Timed outn/an/an/a
Timed outn/an/an/a
Timed outn/an/an/a
Timed outn/an/an/a
Timed outn/an/an/a
Empty answern/an/an/a
Timed outn/an/an/a
Timed outn/an/an/a
Empty answern/an/an/a
OK167.7 s173.0 s2.9 tok/s
OK124.0 s129.3 s3.9 tok/s
Timed outn/an/an/a
OK159.5 s165.3 s3.0 tok/s
OK145.3 s152.0 s3.3 tok/s
OK151.4 s159.8 s3.1 tok/s
OK154.0 s161.9 s3.1 tok/s
Timed outn/an/an/a

Call gpt-oss-20b

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": "openai/gpt-oss-20b",
    "messages": [{"role": "user", "content": "Hello"}],
    "max_tokens": 256,
    "stream": true
  }'

Other models answering

Top of the leaderboard right now.