NVIDIAopen weightsNemotron 3 Super pricing
$0.08 per 1M input tokens and $0.45 per 1M output tokens. In real work that is 0.05¢ per chat answer and 0.13¢ to summarize a 20-page PDF.
What Nemotron 3 Super costs per task
Token counts estimated from a measured model of the same lab (Claude Haiku 5.5), rescaled by tokenizer. Reasoning tokens are billed as output.
| Task | Cost | Tokens used | Price rank | Basis |
|---|---|---|---|---|
| Answer a chat question | 0.05¢$0.46 / 1,000 | 44 in · 774 out · 244 reasoning | 55 of 336 | estimated |
| Reply to a customer email | 0.03¢$0.27 / 1,000 | 265 in · 213 out · 346 reasoning | 74 of 336 | estimated |
| Write a 1,500-word blog post | 0.13¢$1.27 / 1,000 | 66 in · 2,614 out · 200 reasoning | 30 of 336 | estimated |
| Write 10 product descriptions | 0.17¢$1.68 / 1,000 | 333 in · 1,324 out · 2,342 reasoning | 56 of 336 | estimated |
| Translate 1,000 words | 0.08¢$0.82 / 1,000 | 1,345 in · 1,581 out | 53 of 336 | estimated |
| Proofread one page (500 words) | 0.05¢$0.50 / 1,000 | 724 in · 670 out · 304 reasoning | 52 of 336 | estimated |
| Summarize a 20-page PDF | 0.13¢$1.33 / 1,000 | 13,223 in · 493 out · 118 reasoning | 32 of 336 | estimated |
| Extract data from an invoice | 0.05¢$0.49 / 1,000 | 401 in · 421 out · 591 reasoning | 51 of 336 | estimated |
| Analyze a 200-row spreadsheet | ≥0.75¢$7.45 / 1,000 | 3,990 in · 415 out · 15,434 reasoning | 88 of 336 | estimated |
| Answer from 5 retrieved documents (RAG) | 0.10¢$1.01 / 1,000 | 3,807 in · 804 out · 753 reasoning | 54 of 336 | estimated |
| Classify 100 support tickets | 0.07¢$0.71 / 1,000 | 1,283 in · 392 out · 961 reasoning | 17 of 336 | estimated |
| Review 500 lines of code | 0.53¢$5.33 / 1,000 | 5,227 in · 2,392 out · 8,532 reasoning | 96 of 336 | estimated |
| Write a SQL query | 0.07¢$0.65 / 1,000 | 133 in · 616 out · 810 reasoning | 62 of 336 | estimated |
| Run a 50-step coding agent | 20¢$202 / 1,000 | 2,277,775 in · 15,270 out · 29,550 reasoning | 100 of 336 | modeled |
Price at every provider
List prices per 1M tokens (input / output) at each host serving Nemotron 3 Super through OpenRouter today. Hosts can differ in quantization, context and speed.
| Provider | Input / output | Cached input | Context | Uptime |
|---|---|---|---|---|
 DeepInfrabf16 | $0.085 / $0.40 | - | 262K | 100% |
| D DekaLLMfp8 | $0.08 / $0.45 | - | 262K | 100% |
| C Crusoefp8 | $0.30 / $2.40 | $0.15 | 262K | 100% |
Price history
First seen in our price history on Mar 23, 2026 at $0.10 input / $0.50 output. Since then the output price moved from $0.50 to $0.45 per 1M tokens (5 changes).
FAQ
How much does Nemotron 3 Super cost?
Nemotron 3 Super costs $0.08 per 1 million input tokens and $0.45 per 1 million output tokens (list price on Oct 18, 2026). In practice one chat answer costs about 0.05¢ and summarizing a 20-page PDF about 0.13¢.
Where is Nemotron 3 Super cheapest?
Among 3 providers we track, the lowest list price today is at DeepInfra: $0.085 input and $0.40 output per 1M tokens.
Does Nemotron 3 Super charge for reasoning tokens?
Yes. Reasoning (thinking) tokens are billed as output tokens. That is why the cost per task can be much higher than the visible answer suggests - our task costs include them.
Is the Nemotron 3 Super cost per task measured or estimated?
Estimated: Nemotron 3 Super was not run directly. Tokens come from Claude Haiku 5.5 and are rescaled by the tokenizer of NVIDIA. Real usage can differ.