Nemotron 3 Super 120B A12B
Nemotron 3 Super 120B A12BAlternatives?
CheaperClaude Haiku 5.5max×19-87%$0.21Remove sampling parameters+5
BetterGPT-5.6 Lunalow+83%-99%$0.01
OpenQwen3.8 27Bxhigh×6.5-39%$1.01Context: 1M → 262K+5CheaperClaude Opus 5low×16.2-33%$1.10
CheaperGLM 5.3 FlashmaxOpen×15.3-83%$0.28Memory (4-bit): 74 GB → 193 GB+5Claude Haiku 5.5max×19-87%$0.21Remove sampling parameters+5Claude Opus 5low×16.2-33%$1.10GLM 5.3 FlashmaxOpen×15.3-83%$0.28Memory (4-bit): 74 GB → 193 GB+5GPT-6.1 Solmax×10.2-56%$0.72Remove unsupported parameters+5GPT-5.6 Terramax×9.5-15%$1.40Prompts with >272K input tokensGPT-5.6 Lunalow+83%-99%$0.01
MiMo v2.5 ProthinkingOpen×2.8-97%$0.054Memory (4-bit): 74 GB → 614 GB
Muse Glimmer 30BhighOpen×6-96%$0.058Context: 1M → 131K
Gemini 3.1 Flash Litehigh+83%-96%$0.06GPT-6 Lunamax×4.8-96%$0.068Function calling only with none · Chat Completions+3Qwen3.8 27BxhighOpen×6.5-39%$1.01Context: 1M → 262K+5Muse Glimmer 30BhighOpen×6-96%$0.058Context: 1M → 131K
MiniMax M2.7thinkingOpensame+38%$0.30 / $1.20Context: 1M → 205K+1
Mistral Medium 3.5highOpen-50%-69%$0.50Context: 1M → 262K+1Nemotron 3.5 LightningthinkingOpen-67%-95%$0.079Nemotron 3 UltrathinkingOpen×3.3-65%$0.58Memory (4-bit): 74 GB → 336 GBNemotron 3.5 LightningthinkingOpen-67%-95%$0.079Nemotron 3.5 Super VLOpenn/a—$0.19Memory (4-bit): 74 GB → 75 GBNemotron 3.5 Bolt 74B-A9BOpenn/a——Context: 1M → 256KNVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4Openn/a—$0.08 / $0.45Memory (4-bit): not verifiedNVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4Openn/a——NVIDIA Nemotron 3 Ultra 550B A55B NVFP4Openn/a——Context: 1M → 262K+1Nemotron 3 Nano Omni 30B A3B ReasoningOpenn/a—$0 / $0Context: 1M → 256KAll NVIDIA models →Run it yourself
estimateBenchmarks
in Chat & KnowledgeFAQ
How much does Nemotron 3 Super 120B A12B cost?›
$0.13 per 1M input tokens and $1.15 per 1M output tokens (cheapest host: Featherless). A typical benchmark task costs about $1.64 at default effort.
Is Nemotron 3 Super 120B A12B free to use?›
There is no free API; the cheapest paid price is $0.13 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Nemotron 3 Super 120B A12B, and what is it best at?›
#121 of 334 in Chat & Knowledge, #75 of 132 in Coding, #89 of 133 in Agents & Tool Use, #41 of 245 in Writing & Language, #104 of 133 in Vision & Documents. Strongest at: Instructions.
Is Nemotron 3 Super 120B A12B better than MiniMax M2.7?›
On MLCR-AA, Nemotron 3 Super 120B A12B scores 3.3 vs 3.3 — about the same.
How can I use Nemotron 3 Super 120B A12B?›
Through the API of Featherless.
What is the context window of Nemotron 3 Super 120B A12B?›
1M tokens.
When was Nemotron 3 Super 120B A12B released? Is there a newer version?›
Released 2026-03-11. It is the current version in its line.
Is Nemotron 3 Super 120B A12B open source?›
Yes, the weights are open (nvidia-nemotron-open-model-license licence).
Can I run Nemotron 3 Super 120B A12B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Nemotron 3 Super 120B A12B read images?›
No, it accepts text only.
What is Nemotron 3 Super 120B A12B?›
Nemotron 3 Super 120B A12B is an open-weights language model from NVIDIA, released Mar 2026.
What are the best alternatives to Nemotron 3 Super 120B A12B?›
Claude Haiku 5.5 (cheaper), GPT-5.6 Luna (better), Qwen3.8 27B (open weights).