NVIDIA logo

Nemotron 3.5 Lightning (30B A3B)

NVIDIA

Nemotron 3.5 Lightning (30B A3B) is NVIDIA's open-weight mixture-of-experts model with 30 billion total parameters (3B active), built for long-running agentic tasks. It delivers efficient agentic coding performance, ranking second on PinchBench, with a 262K token context window.

Key Specifications

Parameters
30.0B
Context
262.1K
Release Date
August 11, 2026
Average Score
78.5%

Timeline

Key dates in the model's history
Announcement
August 11, 2026
Last Update
August 27, 2026

Technical Specifications

Parameters
30.0B
Training Tokens
-
Knowledge Cutoff
-
Family
-
Capabilities
MultimodalZeroEval

Pricing & Availability

Input (per 1M tokens)
$0.05
Output (per 1M tokens)
$0.20
Max Input Tokens
262.1K
Max Output Tokens
262.1K
Supported Features
Function CallingStructured OutputCode ExecutionWeb SearchBatch InferenceFine-tuning

Benchmark Results

Model performance metrics across various tests and benchmarks

Reasoning

Logical reasoning and analysis
GPQA
NeMo Gym / NeMo Evaluator; BF16; GPQA Diamond (no tools)Self-reported
75.0%

Other Tests

Specialized benchmarks
PinchBench
NeMo Gym / NeMo Evaluator; BF16Self-reported
85.0%
MMLU-Pro
NeMo Gym / NeMo Evaluator; BF16Self-reported
82.0%
IFBench
NeMo Gym / NeMo Evaluator; BF16; looseSelf-reported
72.0%

License & Metadata

License
openmdw_license_v1_1
Announcement Date
August 11, 2026
Last Updated
August 27, 2026

Compare Nemotron 3.5 Lightning (30B A3B)

All comparisons

Similar Models

All Models

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.