Models37in catalog
Largest context1MNemotron 3.5 Lightning (free)

Avg context

278K

Cheapest input

Free/M

Nemotron 3.5 Lightning (free)

Largest output

262K

Fw Nemotron 3 Ultra Nvfp4

Speed tiers

balanced 18fast 18deep 1

Capabilities

What Nvidia models support across the catalog.

Vision

10 / 37

27% of models

Tool use

28 / 37

76% of models

Function calling

28 / 37

76% of models

Extended thinking

28 / 37

76% of models

Streaming

37 / 37

100% of models

Prompt caching

11 / 37

30% of models

All Nvidia models

Sorted by context window, largest first. Tap a row for full specs.

ModelContext
Nemotron 3.5 Lightning (free)1M
Nemotron 3 Ultra1M
Nemotron 3 Ultra (free)1M
Nemotron 3 Ultra (batch)512K
Fw Nemotron 3 Ultra Nvfp4262K
Fw Nemotron Lightning 3 5 30b A3b262K
Gov Nvidia Nemotron Nano 3 30b262K
Nemotron 3.5 Lightning262K
Nemotron 3 Nano 30B A3B262K
Nemotron 3 Nano Omni262K
Nemotron 3 Super262K
Nemotron 3 Super (free)262K
Nemotron 3 Ultra Nvfp4262K
Nemotron Lightning 3p5 30b A3b262K
Nvidia Nemotron 3 5 Lightning262K
Nvidia Nemotron 3 Nano 30b A3b262K
Nvidia Nemotron 3 Super 120b A12b262K
Nvidia Nemotron 3 Ultra 550b A55b262K
Nvidia Nemotron Nano 3 30b262K
Gov Nvidia Nemotron Super 3 120b256K
Nemotron 3 120b A12b256K
Nemotron 3 Nano 30B A3B (free)256K
Nemotron 3 Nano Omni (free)256K
Nvidia Nemotron Super 3 120b256K
Llama 3.1 Nemotron 70B Instruct131K

Showing 25 of 37 models

See rankings

How Nvidia models rank against the full catalog.