BenchLeader
NVIDIAOpen weightsReasoning modelCompare with another model

Nemotron 3.5 Lightning 30B A3B

Nemotron 3.5 Lightning 30B A3B is a NVIDIA open-weights reasoning model, released 11 Aug 2026. It is not yet ranked: 4 independent results so far, and the index needs at least five across two categories. At $0.262 per million tokens blended it is cheaper than most ranked models. Last measured 22 Oct 2026.

Blended price
$0.262/M
$0.184 in · $0.495 out
Output speed
–
First answer
–
Context
262k
Full answer
–
Cost per run
–
Released
11 Aug 2026
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Reasoning53
  2. Coding55
  3. Maths55
  4. Human preference52

Versions

NVIDIA has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.

ModelReleasedIndexRank
Nemotron 3.5 Lightning 30B A3Bthis page11 Aug 2026––
Nemotron Lightning 3.5 30B A3B–––

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmarknot statedSourceTrend
Arena Hard Prompts1375#193Arena (LMArena)
LiveBench 1.0 Reasoningnot in index42.9%#67LiveBench

Coding

Benchmarknot statedSourceTrend
Arena Coding1427#177Arena (LMArena)
LiveBench 1.0 Codingnot in index45.8%#67LiveBench

Agents & tools

Benchmarknot statedSourceTrend
LiveBench Agentic Codingnot in index19.9%#66LiveBench

Maths

Benchmarknot statedSourceTrend
LiveBench Mathematicsnot in index66.5%#67LiveBench
Arena Maths1376#174Arena (LMArena)

Knowledge

Benchmarknot statedSourceTrend
LiveBench 1.0 Data Analysisnot in index43.1%#67LiveBench

Instruction following

Benchmarknot statedSourceTrend
LiveBench Languagenot in index38.3%#67LiveBench
LiveBench Instruction Followingnot in index54.9%#64LiveBench
Arena Instruction Followingnot in index1346#192Arena (LMArena)

Human preference

Benchmarknot statedSourceTrend
Arena Text1353#198Arena (LMArena)
Arena Creative Writingnot in index1288#230Arena (LMArena)
Arena Multi-turnnot in index1349#197Arena (LMArena)
Arena Longer Queriesnot in index1347#203Arena (LMArena)

Composite

Benchmarknot statedSourceTrend
LiveBench 1.0not in index44.5%#67LiveBench

Price history

The price above, per 1M tokens, every day it changed.

$0.00$0.136$0.272$0.408$0.544Sept 26Oct 26
input outputnow $0.184 in · $0.495 out

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.037 per 1M. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.0002$0.0002–
Summarize a 30-page report12,000 / 600$0.0025$0.0012–
Code edit6,000 / 1,500$0.0018$0.0012–
Agentic coding session60,000 / 4,000$0.013$0.0064–
Structured extraction2,000 / 200$0.0005$0.0002–

More from NVIDIA

New this month

See also

Questions people ask

Why does Nemotron 3.5 Lightning 30B A3B have no BenchLeader Index score?
The index needs at least five independent results across at least two categories; as of 25 Oct 2026 Nemotron 3.5 Lightning 30B A3B has 4 that count. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
How much does Nemotron 3.5 Lightning 30B A3B cost per 1M tokens?
As of 25 Oct 2026, Nemotron 3.5 Lightning 30B A3B costs $0.184 per 1M input tokens and $0.495 per 1M output tokens, $0.262 blended (three input tokens per output token). The list price last changed on 21 Oct 2026.
When was Nemotron 3.5 Lightning 30B A3B released?
NVIDIA released Nemotron 3.5 Lightning 30B A3B on 11 Aug 2026.

Data as of 25 Oct 2026. Compare with another model.

Cite as: BenchLeader, “Nemotron 3.5 Lightning 30B A3B: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/nemotron-3-5-lightning-30b-a3b.