1,000 NVIDIA Nemotron-4 340B tokens cost $0.0015 as input or $0.003 as output. Most workloads land in between — a 70/30 input/output split costs about $0.00195. Verified Aug 28, 2026
Quick answer
For NVIDIA Nemotron-4 340B, 1,000 tokens cost $0.0015 as input or $0.003 as output. A representative 70/30 workload is about $0.00195.
Method & trust
The page prices the selected token volume using the model's listed input and output rates, then applies a 70/30 input/output split for the representative workload. Cached input, batch discounts, retries, and provider fees can change the final invoice.
Interactive calculator
1,000 in · 0 out
Input price
$1.50 / 1M tokens
Output price
$3.00 / 1M tokens
Uncached pricing. Add caching and monthly projections in the full calculator.
What 1,000 tokens buys in practice
$0.0018
80% input / 20% output — short replies over conversation history.
$0.00165
90% input / 10% output — long retrieved context, grounded answers.
$0.001575
95% input / 5% output — full documents in, structured extraction out.
Same 1,000 tokens on other models
Related calculations
FAQ
1,000 input tokens cost $0.0015; 1,000 output tokens cost $0.003. A typical 70/30 input/output mix costs about $0.00195. Rates: $1.50/M input, $3.00/M output.
A chat-style request of ~4,800 tokens (4,000 in + 800 out) means 1,000 tokens is roughly 0.2 requests. For a RAG workload with ~8,500 tokens per request, about 0.1 requests.
NVIDIA Nemotron-4 340B has no published cached-input discount on its official pricing page; all input is billed at $1.50/M.