NVIDIA

Llama 3.1 Nemotron Instruct 70B

AI model by NVIDIA. Real-time pricing and benchmark data.

Pricing (per 1M tokens)

Input$1.20
Output$1.20
Blended (3:1)$1.20

Source: Artificial Analysis

Performance

Output Speed108 tok/s
Time to First Token3330ms

Median values from Artificial Analysis

Benchmarks

Editorial Analysis

Profile. Llama 3.1 Nemotron Instruct 70B is a budget-tier option from NVIDIA at $1.20/M output tokens. Despite the low price, it scores competitively on measurable benchmarks — making it a candidate for high-volume production workloads where unit cost dominates.

Latency. On latency, Llama 3.1 Nemotron Instruct 70B reports 108 tok/s output speed and 3330ms time-to-first-token. Output speed is in the normal range — most chat and completion applications will be token-bound, not streaming-bound.

Cost model. For a workload of 1M input tokens + 100K output tokens, Llama 3.1 Nemotron Instruct 70B costs $1.32 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.

Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-08-24.

Compare with similar models

ModelInputOutputSpeed
Llama 3.1 Nemotron Instruct 70BCurrent
$1.20$1.20108 tok/s
GPT-5.6 Luna (low)
$0.20$1.20138 tok/s
MiniMax-M2.5
$0.30$1.2094 tok/s
Solar Pro 4
$0.30$1.2047 tok/s
Qwen3 Next 80B A3B Instruct
$0.15$1.20183 tok/s
MiniMax-M2.1
$0.30$1.2088 tok/s

Example Costs

Single Request
$0.0018
1.0K in / 500 out
1K Requests/day
$1.80
1.0M in / 500.0K out
10K Requests/day
$18.00
10.0M in / 5.0M out