NVIDIA Nemotron 3 Nano 4B

Name: NVIDIA Nemotron 3 Nano 4B
Author: NVIDIA

Catalogue only Open weights

Sovereignty and hosting

Catalogue only

This model is catalogued for reference. It is not hosted on LLM EU yet, and no first-party endpoint serves it here. Requests to it go to the provider directly, under that provider's own terms.

Summary

Nemotron 3 Nano 4B is a 4B-parameter model with a context window of up to 262K tokens, released under the NVIDIA Nemotron Open Model License. Published scores include 91.1 on RULER at 128k, 78.5 on AIME25 and 95.4 on MATH500. The model card lists English as the supported language, so it is not a multilingual option.

Specifications

Provider: NVIDIA (US)
Context window: 262k
Parameters: 3.97B
Modalities: text
License: other · licence
Open weights: yes
Release date: 07 Mar 2026
Tasks: cheap, reasoning, rag, summarization

Languages

Strong: en

Adequate: —

Languages

Strong

Adequate

Not verified: the provider has not documented this claim. — a language is listed as strong only where a source claims it; the card links the evidence where one exists.

When to use it

Strengths

long context for a 4B model
high published maths scores for its size
runs on a single small accelerator

Limits

English only, per the model card
vendor licence rather than an OSI one
benchmark figures come from NVIDIA’s own harness

Pricing

This model is catalogued for reference. It is not hosted on LLM EU yet, and no first-party endpoint serves it here. Requests to it go to the provider directly, under that provider's own terms.

Verification status

Where a field has no source, the card shows it as unverified instead of filling it in. An empty field is information.

LLM EU Verified means documented, not legally certified. It is not an AI Act conformity assessment and LLM EU is not an EU body.

Versions and variants

catalogue · Catalogue only · weights

Example request

curl https://api.llmeu.com/v1/chat/completions \
  -H "Authorization: Bearer $LLMEU_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nemotron-3-nano-4b","messages":[{"role":"user","content":"Hello"}]}'

No playground is deployed for this model: no endpoint serves it, so a request would have nowhere to go.

Benchmarks

RULER (128k): 91.1 (vendor-reported) source
AIME25: 78.5 (vendor-reported) source
MATH500: 95.4 (vendor-reported) source
IFEval-Instruction: 88 (vendor-reported) source

Sources

huggingface.co

Card changelog

07 Mar 2026 — Release date
19 Sept 2026 — Last reviewed

Last reviewed: 19 Sept 2026

Tasks

Low cost