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.
step-5-preview
Unstablestepfun-ai/step-5-preview answered 77.3% of probes on the NVIDIA NIM free API in the last 30 days, with a median time to first token of 17.0 s and 18.0 tok/s. It ranks 8 of 16 tracked models.
Model vs. median
Against the median of models that answered, last 30 days.
step-5-preview median of answering models
Error breakdown
161 failed of 709 probes, last 30 days.
| Failure | Probes | Share of failures |
|---|---|---|
| Empty answer | 111 | 68.9% |
| Network error | 30 | 18.6% |
| Timed out | 17 | 10.6% |
| Not served (404) | 3 | 1.9% |
Time to first token
Each probe. Lower is better.
Throughput
Tokens per second after the first token, each probe. Higher is better.
Availability heatmap
Last 7 days by UTC hour.
all answered some failed none answered not probed
Run history
Last 20 probes.
| Time | Result | First token | Full reply | Speed |
|---|---|---|---|---|
| Timed out | n/a | n/a | n/a | |
| OK | 6.51 s | 16.2 s | 20.7 tok/s | |
| OK | 7.49 s | 25.4 s | 11.2 tok/s | |
| Timed out | n/a | n/a | n/a | |
| Timed out | n/a | n/a | n/a | |
| OK | 111.6 s | 134.4 s | 8.8 tok/s | |
| OK | 293 ms | 7.40 s | 28.1 tok/s | |
| Timed out | n/a | n/a | n/a | |
| Timed out | n/a | n/a | n/a | |
| OK | 107.8 s | 126.5 s | 10.7 tok/s | |
| Timed out | n/a | n/a | n/a | |
| OK | 26.8 s | 57.6 s | 6.5 tok/s | |
| OK | 8.84 s | 18.4 s | 23.9 tok/s | |
| Empty answer | n/a | n/a | n/a | |
| OK | 350 ms | 10.4 s | 19.8 tok/s | |
| OK | 196.0 s | 205.9 s | 2.4 tok/s | |
| OK | 14.2 s | 17.9 s | 28.0 tok/s | |
| OK | 18.1 s | 22.5 s | 22.3 tok/s | |
| OK | 4.48 s | 7.87 s | 63.5 tok/s | |
| OK | 33.5 s | 33.6 s | 14.9 tok/s |
Call step-5-preview
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": "stepfun-ai/step-5-preview",
"messages": [{"role": "user", "content": "Hello"}],
"max_tokens": 256,
"stream": true
}'Other models answering
Top of the leaderboard right now.
- 95 nvidia nemotron-3.5-bolt-e4b 114 ms first token · 86.3 tok/s compare
- 93 nvidia nemotron-3.5-super-vl-120b-a12b 343 ms first token · 126 tok/s compare
- 91 poolside laguna-xs-2.1 175 ms first token · 39.8 tok/s compare
- 82 nvidia nemotron-3-ultra-550b-a55b 1.73 s first token · 48.5 tok/s compare
- 78 nvidia nemotron-3-nano-omni-30b-a3b-reasoning 814 ms first token · 81.8 tok/s compare
- 62 meta muse-glimmer-30b 26.4 s first token · 8.2 tok/s compare