NVIDIA: Nemotron 3 Ultra (free) Free API on OpenRouter

Free API Verified

Available from OpenRouter

★★★★★★★★★★ 3.5 Benchmark-backed score

Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy

Free APIOpenAI compatibleReasoningTool callingTextReasoning

Try in Playground Compare Save API Key

API Details

Copy-ready connection details

Use these values in your client or SDK.

Base URL

https://openrouter.ai/api/v1

Model ID

nvidia/nemotron-3-ultra-550b-a55b:free

API format OpenAI Chat Completions + OpenAI Responses

Technical Details

NVIDIA: Nemotron 3 Ultra (free) specifications

Provider and model catalog

Context window 1.0M

Max output 66K

Status Online

Family nemotron

Released Jun 4, 2026

Last updated Aug 6, 2026

Free listing since Jun 4, 2026

Input text

Output text

Capabilities reasoning, tool calling, temperature control

Open weights Yes

External benchmark references

Benchmark	Score	Metric	Date
SWE-Bench Verified	70.7	resolved	2026-06-04
SWE-Bench Multilingual	67.7	resolve rate	2026-06-04
Terminal-Bench	56.4	success rate	2026-06-04
GPQA	87	accuracy	2026-06-04

AI Recommendation

Should you use NVIDIA: Nemotron 3 Ultra (free)?

NVIDIA: Nemotron 3 Ultra (free) is listed for chat, reasoning workloads and supports a 1.0M context window.

Use it when OpenRouter's free tier is enough for evaluation, demos, or light production traffic.

Best for

Chat
Reasoning

Strengths & Weaknesses

Strengths

Strong reasoning profile
Long context window
Reasoning mode listed
Tool calling support

Watch outs

Free-tier rate limits apply
Vision support is not listed

Benchmark Overview

Benchmark signals for NVIDIA: Nemotron 3 Ultra (free)

Measured data

Intelligence General reasoning and instruction following

37.8/100

Coding Programming and code generation

49.3/100

Agentic Tool use and multi-step tasks

27.4/100

Speed Observed generation speed

101 tok/s

Context Maximum listed context window

1.0M

Pricing

NVIDIA: Nemotron 3 Ultra (free) pricing per 1M tokens

Free tier listed

Input $0 per 1M tokens

Output $0 per 1M tokens

Free access Available OpenRouter

Rate limit 200 req/day (free tier) provider policy

Availability

NVIDIA: Nemotron 3 Ultra (free) availability by provider

2 alternatives

We found 3 provider listings for nemotron-3-ultra-550b-a55b. Check model ID, quota, pricing, and API format before switching providers.

Provider	Model listing	Access	Context	API	Limits
OpenRouter	NVIDIA: Nemotron 3 Ultra (free)	Free tier	1.0M	OpenAI-style	200 req/day (free tier)
Kilo Code	nvidia/nemotron-3-ultra-550b-a55b:free	Free tier	1.0M	OpenAI-style	~200 req/hr
NVIDIA NIM	Nemotron 3 Ultra 550B A55B	Free tier	1.0M	Native	Varies

View OpenRouter setup guide →

Compare

NVIDIA: Nemotron 3 Ultra (free) alternatives to consider

Open comparison

Model	Provider	Context	Access
nvidia/nemotron-3-ultra-550b-a55b:free	Kilo Code	1.0M	Free tier
Nemotron 3 Ultra 550B A55B	NVIDIA NIM	1.0M	Free tier
Ling-3.0-flash (free)	OpenRouter	262K	Free tier
Poolside: Laguna S 2.1 (free)	OpenRouter	262K	Free tier
Poolside: Laguna XS 2.1 (free)	OpenRouter	262K	Free tier

Typical Use Cases

NVIDIA: Nemotron 3 Ultra (free) use cases

Chat

NVIDIA: Nemotron 3 Ultra (free) is tagged for chat in this catalog and works with OpenAI-compatible client libraries.

Reasoning

NVIDIA: Nemotron 3 Ultra (free) is tagged for reasoning in this catalog and works with OpenAI-compatible client libraries.

FAQ

NVIDIA: Nemotron 3 Ultra (free) free API FAQ

Is NVIDIA: Nemotron 3 Ultra (free) free to use?

NVIDIA: Nemotron 3 Ultra (free) is listed with free API access on OpenRouter, subject to the provider's quota and account policy.

What is the NVIDIA: Nemotron 3 Ultra (free) model ID?

The model ID shown in this catalog is nvidia/nemotron-3-ultra-550b-a55b:free.

What are the NVIDIA: Nemotron 3 Ultra (free) free tier rate limits on OpenRouter?

The listed free tier limit is 200 req/day (free tier). Limits can change per account tier, so confirm against the provider dashboard.

What context window does NVIDIA: Nemotron 3 Ultra (free) support?

The listed context window is 1.0M tokens with up to 66K output tokens.

More about NVIDIA: Nemotron 3 Ultra (free)