NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4
Oct 2026
NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4Compare inon
Benchmarks of NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP416
BenchmarkByScoreRank
Knowledge12
Coding3
Multimodal1
No benchmark matches
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGemma 4 31BunstatedOpenwins 54%-7%$0.10 / $0.34Context: 1M → 256K
BetterGPT-6 Lunamaxwins 52%+16%$0.068Function calling only with none · Chat Completions+3
OpenKimi K3maxwins 60%+3,378%$2.00Sampling parameters are fixed+3CheaperGemma 3 27BunstatedOpenwins 44%-42%$0.08 / $0.16Context: 1M → 131K
CheaperQwen 3.5 Flashunstatedwins 42%-34%$0.065 / $0.26ModelScoreCostPer taskWhat changes
Gemma 4 31BunstatedOpenwins 54%-7%$0.10 / $0.34Context: 1M → 256KGemma 3 27BunstatedOpenwins 44%-42%$0.08 / $0.16Context: 1M → 131KQwen 3.5 Flashunstatedwins 42%-34%$0.065 / $0.26ModelScoreCostPer taskWhat changes
GPT-6 Lunamaxwins 52%+16%$0.068Function calling only with none · Chat Completions+3GPT-5.6 Lunaxhighwins 52%+161%$0.085GPT-5.6 Terraxhighwins 55%+2,509%$0.63Prompts with >272K input tokensGPT-6.1 Solmaxwins 59%+2,219%$0.72Remove unsupported parameters+5GPT-6 Solmaxwins 56%+2,219%$1.04ModelScoreCostPer taskWhat changes
Kimi K3maxOpenwins 60%+3,378%$2.00Sampling parameters are fixed+3
GLM 5.2maxOpenwins 59%+1,146%$1.28No image inputGLM 5.3maxOpenwins 59%+1,146%$2.01No image input+4GLM 5unstatedOpenwins 58%+799%$1 / $3.20Context: 1M → 203K+1GLM 5.1unstatedOpenwins 58%+1,146%$1.40 / $4.40Context: 1M → 203K+1ModelScoreCostPer taskWhat changes
NVIDIA Nemotron 3 Ultra 550B A55B NVFP4Opensame——Context: 1M → 262K+1Llama Nemotron Super 49B v1.5unstatedOpenwins 38%—$0.40 / $0.40Context: 1M → 128K+1Llama 3 3 Nemotron Super 49B v1Openwins 37%——Context: 1M → 131K+1NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4Openwins 35%——No image inputLlama 3.1 Nemotron 70B InstructOpenwins 34%—$1.20 / $1.20Context: 1M → 131K+1Nemotron 3 Nano 30B A3BunstatedOpenwins 30%—$0.06 / $0.24Context: 1M → 131K+1Nemotron 4 340B InstructOpenwins 29%—$4.20 / $4.20Context: 1M → 4K+1NVIDIA Nemotron Labs Dory 1.0 4BOpenn/a——Context: 1M → 262K+1All NVIDIA models →Run it yourself
Fits onhardware needs not verified yet
Licenceopenmdw-1.1commercial use with conditions
Benchmarks
all categoriesChat & Knowledge
Arena Text · overall≈1,416#83
Arena Text · Chinese1,460.1#75
Arena Text · English1,427.2#80
Arena Text · French1,462.7#42
Arena Text · instruction following1,397.5#90
Arena Text · non-English1,385#96
Arena Text · Russian1,362.4#112
Arena Text · overall · October 22, 20261,405.4#90
Arena Text · overall · October 15, 20261,408.9#84
Coding
WebDev Arena1,404#66
WebDev Arena · October 23, 20261,398.1#71
WebDev Arena · October 19, 20261,403.8#66
Writing & Language
Arena Text · creative writing≈1,393#74
Vision & Documents
Arena Vision · overall1,204.7#64
FAQ
How much does NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 cost?›
$0.08 per 1M input tokens and $0.45 per 1M output tokens (cheapest host: Pioneer).
Is NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 free to use?›
There is no free API; the cheapest paid price is $0.08 per 1M input tokens. The weights are open, so you can run it yourself.
How good is NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4, and what is it best at?›
#80 of 205 in Chat & Knowledge, #66 of 97 in Coding, #83 of 197 in Writing & Language, #64 of 88 in Vision & Documents.
Is NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 better than Qwen 3?›
On Arena Text · creative writing, NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 scores 1,393 vs 1,393 — about the same.
How can I use NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4?›
Through the NVIDIA API.
What is the context window of NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4?›
1M tokens.
When was NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 released? Is there a newer version?›
Released 2026-10-12. It is the current version in its line.
Is NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 open source?›
Yes, the weights are open (openmdw-1.1 licence).
Can I run NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 read images?›
It accepts text and images as input.
What is NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4?›
NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 is an open-weights language model from NVIDIA, released Oct 2026.
What are the best alternatives to NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4?›
Gemma 4 31B (cheaper), GPT-6 Luna (better), Kimi K3 (open weights).