When was Nemotron 3 Super released?

Nemotron 3 Super was released by NVIDIA on Wednesday, Mar 11 2026.

Who made Nemotron 3 Super?

Nemotron 3 Super was built by NVIDIA. The dominant AI chipmaker, publishing the open Nemotron model family. Ships weights, datasets and training recipes to seed demand for its hardware rather than to sell a chatbot.

What benchmark scores did Nemotron 3 Super get?

Nemotron 3 Super reports 4 tracked benchmark scores — BullshitBench v2: 54%; SWE-Bench Verified: 60.5%; GPQA Diamond: 79.2%; AA Intelligence Index: 36. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.

What is the context window of Nemotron 3 Super?

Nemotron 3 Super has a context window of 1M. That is the maximum amount of input plus output the model can hold in a single request.

How many parameters does Nemotron 3 Super have?

Nemotron 3 Super is reported at 120B parameters.

Is Nemotron 3 Super open source?

Yes. Nemotron 3 Super is open source: the model weights are published together with training code and data under an open license, so it can be downloaded, self-hosted, fine-tuned, and in principle reproduced from scratch.

What came before and after Nemotron 3 Super?

NVIDIA's previous tracked release was Nemotron 3 Nano on Dec 15 2025, 86 days earlier. It was followed by Nemotron 3 Ultra on Jun 4 2026.

Nemotron 3 SuperOpen Source

Released Mar 11 2026

Nemotron 3 Super is an AI model released by NVIDIA on Wednesday, Mar 11 2026, 86 days after Nemotron 3 Nano. It is an open-source model — weights, and training code/data, released under an open license. It is a 120B parameter model with a 1M token context window. Benchmark results (shown below) cover BullshitBench v2, SWE-Bench Verified, GPQA Diamond, and AA Intelligence Index.

Available from


DeepInfra	$0.085	$0.40	256K	bf16
DekaLLM	$0.08	$0.45	256K	fp8

USD per 1M tokens

Benchmarks

Coding

SWE-Bench Verified

60.5%

#47 of 57

Reasoning & science

GPQA Diamond

79.2%

#37 of 59

Knowledge work

AA Intelligence Index

#6 of 6

Robustness

BullshitBench v2

54%

#31 of 94

Compare Nemotron 3 Super with

Nemotron 3 Super

Suggested comparisons

Nemotron 3 SupervsNemotron 3.5 Lightning Nemotron 3 SupervsGPT-6.1 Sol Nemotron 3 SupervsClaude Haiku 5.5 Nemotron 3 SupervsGemini 4 Argon Nemotron 3 SupervsMuse Spark 1.3 Nemotron 3 SupervsGrok 4.7 Nemotron 3 SupervsDeepSeek-V4.1-Flash Nemotron 3 SupervsMistral Large 4 Nemotron 3 SupervsKimi K3 Nemotron 3 SupervsGLM-5.3-Flash Nemotron 3 SupervsQwen3.8-Max-0902

About

Nemotron 3 Super, released March 11, 2026 at GTC, scaled the Nemotron 3 recipe to 120 billion total parameters with about 12 billion active, and stretched the context window to a native 1 million tokens. The pitch was the accuracy-per-second frontier rather than the accuracy leaderboard alone: NVIDIA reported 478 tokens per second on a single B200 and more than twice the throughput of gpt-oss-120b at matched settings, aimed at long-running agents whose cost is measured in wall-clock and watts.

At launch it posted 79.2% on GPQA Diamond, 60.5% on SWE-Bench Verified using the OpenHands scaffold, and a 36 on the Artificial Analysis Intelligence Index — at the time the strongest showing of any US-developed open-weight model, in a class the Chinese open labs had dominated for a year. Like Nano before it, the release included datasets and training recipes alongside the weights.

Frequently asked questions

: Nemotron 3 Super was released by NVIDIA on Wednesday, Mar 11 2026.
: Nemotron 3 Super was built by NVIDIA. The dominant AI chipmaker, publishing the open Nemotron model family. Ships weights, datasets and training recipes to seed demand for its hardware rather than to sell a chatbot.
: Nemotron 3 Super reports 4 tracked benchmark scores — BullshitBench v2: 54%; SWE-Bench Verified: 60.5%; GPQA Diamond: 79.2%; AA Intelligence Index: 36. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.
: Nemotron 3 Super has a context window of 1M. That is the maximum amount of input plus output the model can hold in a single request.
: Nemotron 3 Super is reported at 120B parameters.
: Yes. Nemotron 3 Super is open source: the model weights are published together with training code and data under an open license, so it can be downloaded, self-hosted, fine-tuned, and in principle reproduced from scratch.
: NVIDIA's previous tracked release was Nemotron 3 Nano on Dec 15 2025, 86 days earlier. It was followed by Nemotron 3 Ultra on Jun 4 2026.