Nemotron 3 Ultra

nvidia/nemotron-3-ultra-550b-a55b-20260604

The Nemotron 3 Ultra model from NVIDIA processes and generates text. It has a context length of 512,288 tokens, supports tool use and reasoning, and its weights are publicly available.

  • tool use
  • structured outputs
  • reasoning
  • open-weight

At a glance

providernvidia
releasedJun 4, 2026
context262K
input / M$0.60

Specifications

What the API gives you

Context window
262K tokens
Max output
183K tokens
Input price
$0.60 / 1M tokens
Output price
$2.40 / 1M tokens
Cache read
$0.12 / 1M tokens
Modalities
text → text
Knowledge cutoff
—
Released
Jun 4, 2026

Measured evidence

Public benchmark record

Design Arena Coding
1147 Elo2026-10-21
Design Arena UI
1140 Elo2026-10-21
Design Arena Website
1145 Elo2026-10-21
GPQA Diamond (Epoch AI)
85.4% ± 2.52026-08-10evaluated 2026-08-10
LiveBench Agentic Coding
38.7%2026-10-18
LiveBench Coding
70.7%2026-10-18
LiveBench Data Analysis
54.5%2026-10-18
LiveBench Instruction Following
73.5%2026-10-18
LiveBench Language
70.8%2026-10-18
LiveBench Math
88.7%2026-10-18
LiveBench Reasoning
74.7%2026-10-18
LMArena Agent
-12.32026-10-20
LMArena Agent Steerability
-16.52026-10-20
LMArena Agent Task Outcome
-22.12026-10-20
LMArena Agent Tool Hallucination
-1.12026-10-20
OTIS Mock AIME (Epoch AI)
86.7% ± 5.12026-08-10evaluated 2026-08-10

LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference.

Open-weight model on Hugging Face.

Task evidence

How broad is the support?

Each status is exact to this model version and the selected task. Official release claims remain provisional.

  • Best LLM for Math & ReasoningEstablished evidence

    64% of intended weighted task evidence across 3 independent sources. 7 of 7 configured evaluation feeds currently have results.

  • Best LLM for Logic & Problem SolvingEstablished evidence

    64% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Frontend & UIEstablished evidence

    64% of intended weighted task evidence across 2 independent sources. 6 of 6 configured evaluation feeds currently have results.

  • Best LLM for Competition MathEstablished evidence

    56% of intended weighted task evidence across 2 independent sources. 6 of 6 configured evaluation feeds currently have results.

  • Best LLM for Autonomous AgentsEstablished evidence

    36% of intended weighted task evidence across 2 independent sources. 6 of 6 configured evaluation feeds currently have results.

  • Best LLM for AI AgentsEstablished evidence

    34% of intended weighted task evidence across 2 independent sources. 9 of 9 configured evaluation feeds currently have results.

Show all 38 tasks
  • Best LLM for CodingEstablished evidence

    33% of intended weighted task evidence across 2 independent sources. 7 of 7 configured evaluation feeds currently have results.

  • Best LLM for Agentic CodingEstablished evidence

    26% of intended weighted task evidence across 2 independent sources. 7 of 7 configured evaluation feeds currently have results.

  • Best LLM for Creative WritingProvisional evidence

    44% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Creative Writing or LMArena Text. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for WritingProvisional evidence

    44% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Creative Writing or LMArena Text. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for Code CompletionProvisional evidence

    43% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LMArena WebDev. 3 of 3 configured evaluation feeds currently have results.

  • Best LLM for Fast SummariesProvisional evidence

    42% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Text or Route reliability. 2 of 3 configured evaluation feeds currently have results.

  • Best LLM for a General AssistantProvisional evidence

    41% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from SimpleQA Verified (Epoch AI) or LMArena Instruction Following. 6 of 6 configured evaluation feeds currently have results.

  • Best LLM for Tool & Function CallingProvisional evidence

    38% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Search. 6 of 7 configured evaluation feeds currently have results.

  • Best Open-Weight LLM for Self-HostingProvisional evidence

    36% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Text or Route reliability. 2 of 3 configured evaluation feeds currently have results.

  • Best LLM for Chat & RoleplayProvisional evidence

    33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Creative Writing or LMArena Text. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for Roleplay & CharacterProvisional evidence

    33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Creative Writing or LMArena Text. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for Business WritingProvisional evidence

    33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Creative Writing or LMArena Instruction Following. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for Literary TranslationProvisional evidence

    33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Text or WMT translation. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for Marketing CopyProvisional evidence

    33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Creative Writing or LMArena Instruction Following. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for Everyday TranslationProvisional evidence

    33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Text or WMT translation. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for TranslationProvisional evidence

    33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Text or WMT translation. 3 of 4 configured evaluation feeds currently have results.

  • Best Value LLMProvisional evidence

    33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Text or Route reliability. 2 of 3 configured evaluation feeds currently have results.

  • Best Cheap LLM for High VolumeProvisional evidence

    30% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Text or Route reliability. 2 of 3 configured evaluation feeds currently have results.

  • Cheapest Capable LLMProvisional evidence

    30% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Text or Route reliability. 2 of 3 configured evaluation feeds currently have results.

  • Best LLM for Hermes AgentProvisional evidence

    29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or SWE-rebench. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for OpenHandsProvisional evidence

    29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or SWE-rebench. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for GooseProvisional evidence

    29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or SWE-rebench. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for OpenClawProvisional evidence

    29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or SWE-rebench. 4 of 4 configured evaluation feeds currently have results.

  • Best Local LLM for CodingProvisional evidence

    26% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LMArena WebDev. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for JSON & Schema OutputProvisional evidence

    24% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Text. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for Structured Data ExtractionProvisional evidence

    24% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Text. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for High-Volume ExtractionProvisional evidence

    21% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Route reliability or Structured-output eval. 1 of 3 configured evaluation feeds currently have results.

  • Best LLM for Document SummarizationProvisional evidence

    20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 5 of 6 configured evaluation feeds currently have results.

  • Best LLM for SummarizationProvisional evidence

    20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 5 of 6 configured evaluation feeds currently have results.

  • Best LLM for Debugging & Code ReviewProvisional evidence

    17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LMArena WebDev. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Computer & Browser UseProvisional evidence

    17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or OSWorld Verified. 3 of 3 configured evaluation feeds currently have results.

  • Best LLM for Image UnderstandingProvisional evidence

    11% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Vision or VLMEvalKit tasks. 2 of 3 configured evaluation feeds currently have results.

Recommendation coverage

Where this model appears

Not currently included in a published recommendation. The task evidence panel above says exactly what is confirmed and what is still missing.

Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.