Nemotron
NVIDIA's open models, tuned for throughput rather than benchmarks.
Nemotron is NVIDIA's open-weight family, built to run fast on NVIDIA hardware. The Lightning models activate only a few billion of their parameters per token and are meant for high-throughput agent workloads, and the line includes specialised members like a content-safety classifier rather than one general model stretched over every job. Several are free to call through OpenRouter.
At a glance — Nemotron 3 Ultra
- Cost
- $0.60 in · $2.40 out per 1M tokensMid-rangeFree version too
- Memory
- 262K tokens — about 400 pages
- Reads
- Text
- Can do
- Use tools (agents)
- Structured output
- Thinking mode
- Good at
- Coding: better than 50% · Agentic work: better than 52% · Overall: better than 43%
- Reliability
- 100.00% uptime, last 24h
- Released
- Jun 4, 2026
Advanced details
- API model ID
nvidia/nemotron-3-ultra-550b-a55b- Free version
nvidia/nemotron-3-ultra-550b-a55b:free- Max output
- 183K tokens
- Cached input
- $0.12 per 1M
- Tokenizer
- Other
- Coding index
- 49.3
- Agentic work index
- 20.1
- Overall index
- 22.9
Providers
The same model, hosted by different companies. OpenRouter picks one per request and falls back to the next if it fails.
| Provider | Input /1M | Output /1M | Cached /1M | Max output | Uptime 24h |
|---|---|---|---|---|---|
| DeepInfra · fp4 | $0.50 | $2.20 | $0.10 | 16K | 99.92% |
| BaseTen · fp4 | $0.60 | $2.40 | $0.12 | 183K | 100.00% |
| BaseTen · fp4 | $0.60 | $2.40 | $0.12 | 183K | 99.98% |
| Venice · fp8 | $0.625 | $3.125 | $0.188 | 33K | 96.85% |
Supported parameters
- frequency_penalty
- include_reasoning
- logit_bias
- max_tokens
- min_p
- presence_penalty
- reasoning
- reasoning_effort
- repetition_penalty
- response_format
- seed
- stop
- structured_outputs
- temperature
- tool_choice
- tools
- top_k
- top_p
Live from OpenRouter, refreshed hourly. Scores from Artificial Analysis.
Key info
- Pricing
- Open source
- Category
- Models
- Best for
- Expert
- Made by
- NVIDIA
- Weights
- Open
- Good for
- Throughput, and specialised jobs like safety filtering