Nemotron Mini 4B
A compact open-weight model sized for on-device and edge deployment, where it runs locally with no cloud dependency.
At a glance
Released
Mar 5, 2026
Provider
NVIDIA
Context window
128,000 tokens
Max output
8,000 tokens
Modalities
Text → text
Knowledge cutoff
December 2025
API model ID
nemotron-mini-4bLicense
Open weights (NVIDIA Open Model License)
Pricing
Standard
Input$0.06 / 1M
Output$0.24 / 1M
* Rates are from a third-party aggregator — verify before relying on them. A typical 1,500-token prompt with a 650-token answer costs ≈ $0.0002 on the standard tier. Run your own prompt through the calculator →
Best use cases
On-device assistants
Runs locally on laptops and edge boxes for private, offline use.
Embedded tasks
Handles extraction and short-form generation inside constrained hardware.
Draft-and-verify
Acts as a cheap drafter in speculative decoding with larger Nemotrons.
Strengths & limitations
Strengths
- Tiny and fast
- Runs on edge hardware
- Open weights
Limitations
- Limited reasoning depth
- Small context versus siblings
How to test it
- No-code: open build.nvidia.com (NVIDIA NIM) and run a prompt in the playground.
- API: create a key in build.nvidia.com (NVIDIA NIM) and call the API (example below).
curl https://integrate.api.nvidia.com/v1/chat/completions \
# OpenAI-compatible endpoint
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "nemotron-mini-4b",
"messages": [{"role": "user", "content": "Explain tokenization in one paragraph."}]
}'