NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4
Aug 2026
NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4Compare inon
Benchmarks of NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP418
BenchmarkByScoreRank
Knowledge17
Writing1
No benchmark matches
—per task?
self-host onlyprice
—output speed?
1Mcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-5 Minihighwins 56%—$0.053Context: 1M → 400K
OpenGemma 4 31Bunstatedwins 69%—$0.10 / $0.34Context: 1M → 256K+1BetterGPT-6 Lunamaxwins 68%—$0.068Function calling only with none · Chat Completions+3BetterGPT-5.6 Lunaxhighwins 68%—$0.085BetterGPT-5.6 Terraxhighwins 70%—$0.63Prompts with >272K input tokensModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-5 Minihighwins 56%—$0.053Context: 1M → 400KGPT-6 Lunamaxwins 68%—$0.068Function calling only with none · Chat Completions+3GPT-5.6 Lunaxhighwins 68%—$0.085GPT-5.6 Terraxhighwins 70%—$0.63Prompts with >272K input tokensGPT-6.1 Solmaxwins 73%—$0.72Remove unsupported parameters+5ModelScoreCostPer taskWhat changes
Gemma 4 31BunstatedOpenwins 69%—$0.10 / $0.34Context: 1M → 256K+1Gemma 4 26B A4BunstatedOpenwins 66%—$0.15 / $0.60Context: 1M → 256K+1
Qwen3.8 27BunstatedOpenwins 62%—$0.42 / $2.55Context: 1M → 262K+6
Muse Glimmer 30BunstatedOpenwins 61%—$0.30 / $1.10Context: 1M → 131K+1Qwen 3.5 27BunstatedOpenwins 61%—$0.30 / $2.40Context: 1M → 262K+1ModelScoreCostPer taskWhat changes
NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4Openwins 65%—$0.08 / $0.45Memory (4-bit): not verifiedNVIDIA Nemotron 3 Ultra 550B A55B NVFP4Openwins 64%——Context: 1M → 262K+1Llama Nemotron Super 49B v1.5unstatedOpenwins 53%—$0.40 / $0.40Context: 1M → 128K+1Llama 3 3 Nemotron Super 49B v1Opensame——Context: 1M → 131K+2Llama 3.1 Nemotron 70B InstructOpensame—$1.20 / $1.20Context: 1M → 131K+1Nemotron 3 Nano 30B A3BunstatedOpenwins 44%—$0.06 / $0.24Context: 1M → 131K+1Nemotron 4 340B InstructOpenwins 44%—$4.20 / $4.20Context: 1M → 4K+1NVIDIA Nemotron Labs Dory 1.0 4BOpenn/a——Context: 1M → 262KAll NVIDIA models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✓ 16 GB · compressed✓ 24 GB GPU · 8-bit✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
Licenceopenmdw-1.1commercial use with conditions
Benchmarks
all categoriesChat & Knowledge
Arena Text · overall1,351#123
Arena Text · Chinese1,363.6#132
Arena Text · English1,367.7#129
Arena Text · French1,379.9#110
Arena Text · German1,289.9#123
Arena Text · instruction following1,320.5#130
Arena Text · Japanese1,222.7#116
Arena Text · Korean1,241.8#121
Arena Text · non-English1,297.3#147
Arena Text · Polish1,278.8#123
Arena Text · Russian1,254.9#160
Arena Text · Spanish1,358.4#115
Arena Text · overall · October 22, 20261,330.5#136
Arena Text · overall · October 15, 20261,328.9#133
Writing & Language
Arena Text · creative writing1,283#142
EQ-Bench Creative Writing v31,280.8#69
FAQ
How much does NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 cost?›
There is no per-unit API price: self-host only.
Is NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4, and what is it best at?›
#129 of 205 in Chat & Knowledge, #148 of 197 in Writing & Language.
Is NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 better than MiniMax M2?›
On Arena Text · creative writing, NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 scores 1,283 vs 1,283 — about the same.
What is the context window of NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4?›
1M tokens.
When was NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 released? Is there a newer version?›
Released 2026-08-04. It is the current version in its line.
Is NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 open source?›
Yes, the weights are open (openmdw-1.1 licence).
Can I run NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 read images?›
No, it accepts text only.
What is NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4?›
NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4 is an open-weights language model from NVIDIA, released Aug 2026.
What are the best alternatives to NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4?›
GPT-5 Mini (better), Gemma 4 31B (open weights).