Nemotron 3 Ultra
Nemotron 3 UltraAlternatives?
CheaperQwen3.8 FlashthinkingOpenwins 98%-35%$0.37Context: 1M → 262K+6
BetterGPT-6 Lunanon-thinkingwins 54%-98%$0.011Function calling only with none · Chat Completions+3
CheaperClaude Haiku 5.5maxwins 98%-63%$0.21Remove sampling parameters+5
CheaperDeepSeek V4.1 Flash 1017maxOpenwins 98%-55%$0.26Memory (4-bit): 336 GB → 458 GB+5
CheaperMiMo v2.6 ProthinkingOpenwins 98%-77%$0.13Memory (4-bit): 336 GB → 615 GB+4Qwen3.8 FlashthinkingOpenwins 98%-35%$0.37Context: 1M → 262K+6Claude Haiku 5.5maxwins 98%-63%$0.21Remove sampling parameters+5DeepSeek V4.1 Flash 1017maxOpenwins 98%-55%$0.26Memory (4-bit): 336 GB → 458 GB+5MiMo v2.6 ProthinkingOpenwins 98%-77%$0.13Memory (4-bit): 336 GB → 615 GB+4GPT-6.1 Solxhighwins 97%-32%$0.39Remove unsupported parameters+5GPT-6 Lunanon-thinkingwins 54%-98%$0.011Function calling only with none · Chat Completions+3GPT-5.6 Lunamediumwins 58%-97%$0.016Claude Haiku 5.5lowwins 79%-96%$0.024Remove sampling parameters+5MiMo v2.6 FlashthinkingOpenwins 97%-89%$0.062Pass back reasoning_content+3
Hy3thinkingOpenwins 60%-88%$0.072Context: 1M → 256KQwen3.8 FlashthinkingOpenwins 98%-35%$0.37Context: 1M → 262K+6DeepSeek V4.1 Flash 1017maxOpenwins 98%-55%$0.26Memory (4-bit): 336 GB → 458 GB+5MiMo v2.6 ProthinkingOpenwins 98%-77%$0.13Memory (4-bit): 336 GB → 615 GB+4
GLM 5.3maxOpenwins 97%+248%$2.01Memory (4-bit): 336 GB → 452 GB+4MiMo v2.6 FlashthinkingOpenwins 97%-89%$0.062Pass back reasoning_content+3Nemotron 3.5 Super VLOpenwins 67%-67%$0.19Nemotron 3.5 Bolt 74B-A9BOpenwins 55%——Context: 1M → 256KNemotron 3.5 LightningthinkingOpenwins 11%-86%$0.079Nemotron 3 Nano 30B A3BthinkingOpenwins 1%-97%$0.017Context: 1M → 131KNVIDIA Nemotron 3 Nano 4BOpenwins 1%——Context: 1M → 262KNemotron 3 Super 120B A12BOpenwins 1%+184%$1.64Nemotron 3 Nano Omni 30B A3B ReasoningOpenn/a—$0 / $0Context: 1M → 256KNemotron Cascade 2 30B A3BOpenn/a——All NVIDIA models →Run it yourself
estimateBenchmarks
in Agents & Tool UseFAQ
How much does Nemotron 3 Ultra cost?›
$0.50 per 1M input tokens and $2.50 per 1M output tokens; cached input $0.15 per 1M. A typical benchmark task costs about $0.58 at default effort.
Is Nemotron 3 Ultra free to use?›
There is no free API; the cheapest paid price is $0.50 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Nemotron 3 Ultra, and what is it best at?›
#90 of 334 in Chat & Knowledge, #63 of 132 in Coding, #92 of 133 in Agents & Tool Use, #3 of 245 in Writing & Language, #87 of 133 in Vision & Documents. Strongest at: Science, Long context, Knowledge, Instructions.
Is Nemotron 3 Ultra better than MiMo v2.5 Pro?›
On AA-Briefcase 1.1 · Elo, Nemotron 3 Ultra scores 876.4 vs 882.8 — about the same. It is 967% more expensive per task.
How can I use Nemotron 3 Ultra?›
Through the API of BaseTen, DeepInfra, Venice.
What is the context window of Nemotron 3 Ultra?›
1M tokens, with up to 66K output tokens.
When was Nemotron 3 Ultra released? Is there a newer version?›
Released 2026-06-04. It is the current version in its line.
Is Nemotron 3 Ultra open source?›
Yes, the weights are open (openmdw-1.1 licence).
Can I run Nemotron 3 Ultra locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Nemotron 3 Ultra read images?›
No, it accepts text only.
What is Nemotron 3 Ultra?›
Nemotron 3 Ultra is an open-weights language model from NVIDIA, released Jun 2026.
What are the best alternatives to Nemotron 3 Ultra?›
Qwen3.8 Flash (cheaper), GPT-6 Luna (better).