Nemotron 4 340B Instruct pricing

Cheapest host: DeepInfra · checked Oct 26, 2026
Benchmarks and alternatives →
$4.20 / $4.20per 1M tokens in / out

Price sheetper 1M tokens

Input$4.20DeepInfra
Output$4.20
Context4Ktokens

Where to buy1 provider · per 1M tokens

ProviderInputOutputContext
DeepInfra$4.20$4.20—

In NVIDIA's line-upper 1M in / out

NVIDIA Nemotron Nano 12B v2 VL$0.20 / $0.60−93%
Nemotron 3 Super 120B A12B$0.13 / $1.15−91%
Nemotron 3 Ultra$0.60 / $2.40−75%
Llama 3.1 Nemotron 70B Instruct$1.20 / $1.20−71%
OpenCodeReasoning Nemotron 1.1 32B$2.72 / $4.80−23%
OpenReasoning Nemotron 32B$2.72 / $4.80−23%
Nemotron 4 340B Instruct$4.20 / $4.20this model
<

Cheaper for the same workvs Nemotron 4 340B Instruct

MiMo v2.6 Pro unstatedOpen-87%score wins 82%$0.43 / $0.87
Gemini 3.8 Flash high-64%score wins 82%$3.49
Muse Spark 1.3 max-52%score wins 82%$3.69
Muse Spark 1.2 xhigh-52%score wins 81%$1.29
GLM 5.3 maxOpen-49%score wins 80%$2.01

Pricing questions

How much does Nemotron 4 340B Instruct cost?›

$4.20 per 1M input tokens and $4.20 per 1M output tokens.

Is Nemotron 4 340B Instruct free to use?›

There is no free API tier, but the weights are open: you can run it on your own hardware for the cost of compute.

What is a cheaper alternative to Nemotron 4 340B Instruct?›

MiMo v2.6 Pro at unstated effort costs -87% per task with a score wins 82% on Arena Text · English.