NVIDIAReleased Aug 11, 2026
Open weights

Nemotron 3.5 Lightning

Its best results are in Knowledge Work (#119 of 186) and Academic Reasoning (#161 of 223); weakest in Programmatic Reasoning (#166 of 187). Three of seven dimensions are estimated, because no results are published yet.

Composite IQ ?
100
#136 of 186 ranked models · ahead of 27%
US (Other)'s #10 model, behind Inkling 2 Small (125)
Effective cost
$0.269/1M
$0.07 in · $0.22 out
Speed
308tok/s
9s typical response
Context
1Mtokens
1,000,000 tokens in
Evidence
4of 7 dims
12 of 39 benchmarks

IQ profile

Seven dimensions, averaged equally into Composite IQ
Compare shape with
100 avg human130160● Abstract84#158 · est.● Math112#91 · est.● Academic104#161● Coding94#166● Computer Use97#136 · est.● Reliability104#145● Knowledge Work102#119
MeasuredEstimated95% range
#1 in dimensionTop-10 median

IQ dimensions

Ranks are among models scored in each dimension. Ranges are model-based estimates, not calibrated confidence intervals.

DimensionIQPosition · 70–160RankEvidencevs Top 10
84estimated
#158 / 221
0/5
-48.5
112estimated
#91 / 186
0/5
-37.5
104measured
#161 / 223
4/6
-31
94measured
#166 / 187
2/6
-57
97estimated
#136 / 186
0/5
-49.5
104measured
#145 / 188
2/6
-31.5
102measured
#119 / 186
4/6
-30
Nemotron 3.5 Lightning95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

12 of 39 Composite IQ benchmarks have a published result for Nemotron 3.5 Lightning. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 84 · estimatedNo published results0/5
No published results yet for Abstract Reasoning. The 84 shown is estimated from Nemotron 3.5 Lightning's other dimensions and will be replaced as results come in. How estimates work
ARC-AGI-3ARC-AGI-2ARC-AGI-1Chess PuzzlesMystery Game Puzzles
Mathematical Reasoning IQ 112 · estimatedNo published results0/5
No published results yet for Mathematical Reasoning. The 112 shown is estimated from Nemotron 3.5 Lightning's other dimensions and will be replaced as results come in. How estimates work
FrontierMath Tier 4FrontierMath Tier 1-3ProofBench 1.1MathArenaAIME
Academic Reasoning IQ 104 · measuredBest: #123 of 147 on GPQA Diamond4/6
BenchmarkScorePercentileRankLeader · source
GPQA Diamond
Artificial Analysis v5.0.1 · Nov 13
74.3%
#123 / 147
Leader gpt-6-astra · 96.1%
Artificial Analysis v5.0.1 · Nov 13
Humanity's Last Exam
Artificial Analysis v4.3.2 · Oct 27
10.6%
#141 / 167
Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 27
SciCode
Artificial Analysis v4.3.2 · Oct 27
32.1%
#149 / 167
Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 27
CritPt
Artificial Analysis v4.3.2 · Oct 27
0.0%
#154 / 159
Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 27
Not yet reported
MMLU-ProMMMU-Pro
Programmatic Reasoning IQ 94 · measuredBest: #72 of 109 on Terminal-Bench 4.02/6
BenchmarkScorePercentileRankLeader · source
Terminal-Bench 4.0
Artificial Analysis v5.0.1 · Nov 13
0.5%
#72 / 109
Leader Kimi K3.1 Pro · 67.2%
Artificial Analysis v5.0.1 · Nov 13
ProgramBench
Full source leaderboard reconciliation · Oct 23
0.0%
#63 / 63
Leader opus-5.5 · 65.0%
Full source leaderboard reconciliation · Oct 23
Not yet reported
LiveCodeBenchIOI V2SWE-rebenchMirrorCode
Computer Use IQ 97 · estimatedNo published results0/5
No published results yet for Computer Use. The 97 shown is estimated from Nemotron 3.5 Lightning's other dimensions and will be replaced as results come in. How estimates work
BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam
Reliability IQ 104 · measuredBest: #108 of 158 on AA Omniscience2/6
BenchmarkScorePercentileRankLeader · source
AA Omniscience
Artificial Analysis v4.3.2 · Oct 27
-17.7
#108 / 158
Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 27
AA Long Context Reasoning v1.1
Artificial Analysis v4.3.2 · Oct 27
60.3%
#128 / 154
Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 27
Not yet reported
SimpleQA VerifiedBullshitBench v2IFBenchFACTS Grounding
Knowledge Work IQ 102 · measuredBest: #97 of 124 on GDPval-AA v2.1 normalized score4/6
BenchmarkScorePercentileRankLeader · source
GDPval-AA v2.1 normalized score
Artificial Analysis v4.3.2 · Oct 23
6.8%
#97 / 124
Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 23
AA-Briefcase v1.1 Elo
Artificial Analysis v5.0.1 · Nov 13
507
#93 / 116
Leader sonnet-5.5 · 1,816
Artificial Analysis v5.0.1 · Nov 13
GDP.pdf
Artificial Analysis v5.0.1 · Nov 13
3.0%
#96 / 111
Leader Muse Wonder Full · 32.6%
Artificial Analysis v5.0.1 · Nov 13
AutomationBench-AA
Artificial Analysis v5.0.1 · Nov 13
0.8%
#104 / 112
Leader Gemini 4 Carbon · 77.8%
Artificial Analysis v5.0.1 · Nov 13
Not yet reported
AA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released
Developer
NVIDIA
Access
Open weights for local deployment and hosted access through model providers
Weights
Open weightsDownloadable and self-hostable
Inputs
text
Context
1,000,000 tokens
API price
$0.07 in · $0.22 outper 1M tokens · full pricing

What NVIDIA says

  • The hybrid mixture-of-experts model has 31.6B total parameters and 3.6B active parameters.
  • NVIDIA designed it for efficient reasoning, coding, and agentic workloads with up to 1M tokens of context.

Limitations

  • The model is text-only.

Publisher statements, not AI IQ measurements.

About this score

How much of Nemotron 3.5 Lightning's IQ is measured, and how the rest is estimated.

12/ 39benchmark results
4/ 7dimensions measured
3estimated
0/5Abstract
0/5Math
4/6Academic
2/6Coding
0/5Computer Use
2/6Reliability
4/6Knowledge Work
Bar height = share of benchmarks with resultsEstimatedNo score
Measured results are shown as published

Every score in the benchmark tables comes from its source unchanged. Composite IQ averages all seven dimensions equally.

Estimates fill the gaps

Abstract Reasoning, Mathematical Reasoning and Computer Use have no published results yet, so their IQ is estimated from Nemotron 3.5 Lightning's other dimensions. Missing benchmarks can also draw on its predecessor, Nemotron 3 Nano; a predecessor is an estimation input, not evidence that this model has the predecessor's measured results.

A missing result is not a zero

Benchmark versions and evaluation harnesses can differ between sources, so small gaps between models may not be meaningful.

Read the scoring methodology →

Compare Nemotron 3.5 Lightning

Side-by-side pages for the models it is most often weighed against.

Nemotron 3.5 LightningComparison model
Browse all comparisons →

Frequently asked questions

Quick answers about Nemotron 3.5 Lightning, drawn from the same data as the rest of this page.

How AI IQ scores models →Get this data from the API →
What is the IQ of Nemotron 3.5 Lightning?
Nemotron 3.5 Lightning has an estimated IQ of 100, which ranks #136 out of 186 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.
What is Nemotron 3.5 Lightning best at?
Among the dimensions with published benchmark results, Nemotron 3.5 Lightning ranks highest in Knowledge Work (IQ 102, #119 of 186) and Academic Reasoning (IQ 104, #161 of 223). Its lowest-ranked measured dimension is Programmatic Reasoning (IQ 94, #166 of 187).
What is the EQ of Nemotron 3.5 Lightning?
Nemotron 3.5 Lightning doesn't have an EQ score yet. EQ (emotional reasoning) is calculated from EQ-Bench 4 and AttuneBench, and neither has published a result for Nemotron 3.5 Lightning. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has Nemotron 3.5 Lightning been tested on?
Nemotron 3.5 Lightning has published results on 12 of the 39 benchmarks behind AI IQ's Composite IQ, including Terminal-Bench 4.0, AA Omniscience, GDPval-AA v2.1 normalized score, AA-Briefcase v1.1 Elo and AA Long Context Reasoning v1.1.
How much does Nemotron 3.5 Lightning cost?
Nemotron 3.5 Lightning costs $0.07 per million input tokens and $0.22 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.2689 per million input-plus-output tokens.
Is Nemotron 3.5 Lightning open source?
Yes — Nemotron 3.5 Lightning is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed Nemotron 3.5 Lightning?
Nemotron 3.5 Lightning was developed by NVIDIA, based in the United States.
When was Nemotron 3.5 Lightning released?
Nemotron 3.5 Lightning was publicly released on August 11, 2026.
How large is the context window of Nemotron 3.5 Lightning?
Nemotron 3.5 Lightning supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.
How fast is Nemotron 3.5 Lightning?
In independent measurements, Nemotron 3.5 Lightning generates about 308 tokens per second, with a typical response time of about 8.7 seconds.