Nemotron 3 Ultra
NVIDIA's 550B/55B-active frontier reasoning model for demanding agent workloads.
- Modalities
- IntextOuttext
- Input / output
- $0.90 / $2.40USD per 1M tokens
- Context
- 262K
- Providers
- 4 live
- Released
- Jun 4, 2026
- Knowledge cutoff
- September 2025
Playground
Preparing playground
Performance
Response latency estimates
Time to first token
Generation speed · tok/s
Request success
Text-model timing and speed, plus API request success. Daily latency percentiles are estimates. Your request logs
API
POST/api/v1/chat/completionsOpenAI-compatible
nemotron-3-ultra
curl https://www.ninjachat.ai/api/v1/chat/completions \
-H "Authorization: Bearer $NINJACHAT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "nemotron-3-ultra",
"messages": [{ "role": "user", "content": "Hello!" }],
"stream": true
}'messagestemperaturemax_completion_tokenstop_pstopfrequency_penaltypresence_penaltyseedstreamuserroutingtoolstool_choiceresponse_formatreasoningreasoning_effort
Pricing
- Input
- $0.90/M tokens
- Cached input
- $0.09/M tokens
- Output
- $2.40/M tokens
Published USD rates. Actual cost depends on input, output, cache usage and request options.
