NVIDIA Nemotron 3 Super 120B A12B
3130°
nvidia | 262,144 token context | $0.09/M input, $0.40/M output. NVIDIA Nemotron 3 Super is a hybrid Mixture-of-Experts (MoE) model engineered for highest compute efficiency and accuracy in multi-agent applications and specialized agentic systems. It is optimized to run many collaborating agents per application on a single GPU, delivering high accuracy for reasoning, tool use, and instruction following.
Guides for this model
Open weights vs open source vs API-only: what you are actually allowed to do
How to choose an AI model in October 2026: the closed source news, then the open weights worth downloading
What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)
Captain's notes
Notes from 2026-10-13
NVIDIA Nemotron Super 120B A12B is a hybrid Mixture-of-Experts model from NVIDIA built specifically for multi-agent and agentic workloads. Its MoE architecture activates only 12B parameters at a time, letting you run many agents in parallel on a single GPU without the memory cost of a dense 120B model. If you are orchestrating tool-calling pipelines or agent swarms, this is the efficiency angle NVIDIA is selling.
Known for
- MoE design activates ~12B parameters per forward pass, cutting compute per agent
- Built for multi-agent pipelines: tool use, instruction following, and agent coordination
- 262,144-token context window handles long task chains and large documents
- Priced well below most frontier models at $0.09/M input and $0.40/M output
Watch out for
- Weights are not public; you are API-only with no option to self-host or inspect the model.
- Agentic use cases can rack up output tokens fast — watch the $0.40/M output cost in long pipelines.
- No model card details published, so independent benchmark verification is not possible.
Prompts to try
Multi-step tool-use plan
You are coordinating a research agent. Given the goal: 'Find the three most-cited papers on transformer attention efficiency published after 2022 and summarize their key contributions', break this into discrete tool calls (web search, paper fetch, summarize), specify the input and expected output of each step, and flag any step where a fallback is needed if the tool returns no results.
Agent handoff protocol
Design a handoff message format for a customer-support agent swarm. There are three agents: Triage, Billing, and Technical. Write the JSON schema for a handoff payload, then write one example handoff from Triage to Billing for a customer who was charged twice for the same subscription and is threatening to cancel.
Long-context instruction chain
Here is a 2,000-word product requirements document. Read it carefully, then produce: (1) a numbered list of every explicit acceptance criterion, (2) a list of any requirements that contradict each other, and (3) three clarifying questions the engineering team should ask before starting work. [Paste your PRD below this line.]
Compare with
- Nemotron 3 UltraOpen-weights model262,144 token context · $0.60/M input, $2.40/M output
- Nemotron 3.5 LightningOpen-weights model262,144 token context · $0.07/M input, $0.20/M output
- Nemotron 3 SuperOpen-weights model262,144 token context · $0.08/M input, $0.45/M output
- NVIDIA Nemotron 3 Ultra 550B A55BOpen-weights model262,144 token context · $0.50/M input, $2.20/M output
More text models from nvidia
Questions people ask
- Can I run Nemotron Super 120B locally?
- No. NVIDIA has not released open weights for this model, so it is API-only and cannot be self-hosted.
- What does it cost and how long is the context?
- Pricing is $0.09 per million input tokens and $0.40 per million output tokens, with a 262,144-token context window.
- How is this different from a standard 120B dense model?
- The MoE architecture means only around 12B parameters are active per inference step, so it is far cheaper to run at scale and can fit more simultaneous agent instances on a single GPU than a dense model of the same nominal size.
Captain's verdictA sensible pick for agent pipeline work where you need long context and low per-call cost, but you are fully vendor-locked to NVIDIA's API.
Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.
Ranking badge
Updated Oct 22, 2026
Made this model? Show this badge on your page.
Paste it into your Hugging Face card or Civitai description.
How this score adds up
This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.
Glossary: the terms used here
- Heat score
- One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
- Floor
- The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
- Download-equivalents
- Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
- Log scale
- Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
- Likes
- Thumbs up and likes from every source, worth half a download each on the log scale.
- Adoption rate
- How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
- Measured and estimated
- Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
- Seeder
- Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
- AI Horde
- A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
- Generation sites
- Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
- OpenRouter tokens
- A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
- Full model (checkpoint)
- The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
- LoRA and embedding
- Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
- Family and weights
- The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
- Age
- Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
- Family boost
- A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
- Stand-in score
- A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
- New! (grace period)
- A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
- Trend
- How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
- Typical model
- The middle of the catalog: half the models with a figure grew faster over the period, half slower.
NVIDIA Nemotron 3 Super 120B A12B is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.
Browse by Model Family
Text / LLM Models
434
Anima Models
528
Chroma Models
19
Fibo Models
2
Flux Models
1,496
Flux 2 / Klein Models
1,239
Flux 3 Models
3
Moss Models
3
Iris Models
1
MiniMax H3 Models
364
Hunyuan Models
204
Ideogram Models
20
Krea2 Models
717
Qwen Models
20
Qwen2 Models
47
Ltx2 Models
153