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.
llama-3.2-90b-vision-instruct
Unstablemeta/llama-3.2-90b-vision-instruct answered 74.7% of probes on the NVIDIA NIM free API in the last 7 days, with a median time to first token of 9.73 s and 4.8 tok/s. It ranks 9 of 15 tracked models.
Model vs. median
Against the median of models that answered, last 7 days.
llama-3.2-90b-vision-instruct median of answering models
Error breakdown
40 failed of 158 probes, last 7 days.
| Failure | Probes | Share of failures |
|---|---|---|
| Not served (404) | 21 | 52.5% |
| Timed out | 14 | 35.0% |
| Empty answer | 4 | 10.0% |
| Server error (5xx) | 1 | 2.5% |
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 |
|---|---|---|---|---|
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a | |
| Not served (404) | n/a | n/a | n/a |
Call llama-3.2-90b-vision-instruct
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": "meta/llama-3.2-90b-vision-instruct",
"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
- 88 nvidia nemotron-3-nano-omni-30b-a3b-reasoning 731 ms first token · 88.0 tok/s compare
- 79 nvidia nemotron-3-ultra-550b-a55b 1.75 s first token · 48.0 tok/s compare
- 62 meta muse-glimmer-30b 26.4 s first token · 8.2 tok/s compare