Llama 3 3 Nemotron Super 49B v1
Mar 2025
Llama 3 3 Nemotron Super 49B v1Compare inon
Benchmarks of Llama 3 3 Nemotron Super 49B v12
BenchmarkByScoreRank
No benchmark matches
—per task?
self-host onlyprice
—output speed?
131Kcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-5 Minihighwins 54%—$0.053
OpenGemma 4 31Bunstatedwins 67%—$0.10 / $0.34BetterGPT-6 Lunamaxwins 66%—$0.068Function calling only with none · Chat Completions+3BetterGPT-5.6 Lunaxhighwins 66%—$0.085BetterGPT-5.6 Terraxhighwins 68%—$0.63Prompts with >272K input tokensModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-5 Minihighwins 54%—$0.053GPT-6 Lunamaxwins 66%—$0.068Function calling only with none · Chat Completions+3GPT-5.6 Lunaxhighwins 66%—$0.085GPT-5.6 Terraxhighwins 68%—$0.63Prompts with >272K input tokensGPT-6.1 Solmaxwins 72%—$0.72Remove unsupported parameters+5ModelScoreCostPer taskWhat changes
Gemma 4 31BunstatedOpenwins 67%—$0.10 / $0.34Gemma 4 26B A4BunstatedOpenwins 64%—$0.15 / $0.60
Qwen3.8 27BunstatedOpenwins 60%—$0.42 / $2.55max_tokens range · QwenCloud+4
Muse Glimmer 30BunstatedOpenwins 60%—$0.30 / $1.10Qwen 3.5 27BunstatedOpenwins 59%—$0.30 / $2.40ModelScoreCostPer taskWhat changes
NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4Openwins 63%—$0.08 / $0.45Memory (4-bit): not verifiedNVIDIA Nemotron 3 Ultra 550B A55B NVFP4Openwins 62%——Memory (4-bit): 30 GB → 182 GBLlama Nemotron Super 49B v1.5unstatedOpensame—$0.40 / $0.40Context: 131K → 128KNVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4Opensame——Llama 3.1 Nemotron 70B InstructOpenwins 47%—$1.20 / $1.20Memory (4-bit): not verifiedNemotron 3 Nano 30B A3BunstatedOpenwins 42%—$0.06 / $0.24Nemotron 4 340B InstructOpenwins 42%—$4.20 / $4.20Context: 131K → 4K+1Nemotron 3.5 Super VLOpenn/a—$0.19Memory (4-bit): 30 GB → 75 GBAll NVIDIA models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✓ 48 GB GPU · compressed✓ 80 GB GPU · 8-bit
Licencenvidia-open-model-licensecommercial use with conditions
History
2 versionsBenchmarks
all categoriesChat & Knowledge
Arena Text · overall1,328#138
Writing & Language
Arena Text · creative writing1,297#132
FAQ
How much does Llama 3 3 Nemotron Super 49B v1 cost?›
There is no per-unit API price: self-host only.
Is Llama 3 3 Nemotron Super 49B v1 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Llama 3 3 Nemotron Super 49B v1, and what is it best at?›
#138 of 196 in Chat & Knowledge, #132 of 195 in Writing & Language.
Is Llama 3 3 Nemotron Super 49B v1 better than Gemma 3n E4B Instruct?›
On Arena Text · creative writing, Llama 3 3 Nemotron Super 49B v1 scores 1,297 vs 1,299 — about the same.
What is the context window of Llama 3 3 Nemotron Super 49B v1?›
131K tokens.
When was Llama 3 3 Nemotron Super 49B v1 released? Is there a newer version?›
Released 2025-03-16. It is the current version in its line.
Is Llama 3 3 Nemotron Super 49B v1 open source?›
Yes, the weights are open (nvidia-open-model-license licence).
Can I run Llama 3 3 Nemotron Super 49B v1 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Llama 3 3 Nemotron Super 49B v1 read images?›
No, it accepts text only.
What is Llama 3 3 Nemotron Super 49B v1?›
Llama 3 3 Nemotron Super 49B v1 is an open-weights language model from NVIDIA, released Mar 2025.
What are the best alternatives to Llama 3 3 Nemotron Super 49B v1?›
GPT-5 Mini (better), Gemma 4 31B (open weights).