NVIDIA Nemotron Nano 9B V2
Aug 2025
NVIDIA Nemotron Nano 9B V2Compare inon
Benchmarks of NVIDIA Nemotron Nano 9B V213
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Documents1
Legacy · saturated or no longer reported4
No benchmark matches
Rankingin thinking mode?
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperQwen3.5 4BthinkingOpen+77%-14%$0.03 / $0.15
BetterGPT-6 Lunalow×2.9+186%$0.005Function calling only with none · Chat Completions+3IOpenK2 Horizon 7Bthinking×2.8+650%$0.30 / $1.20Memory (4-bit): 4.4 GB → 4.5 GBCheaperNemotron 3.5 LightningthinkingOpen+73%-30%$0.079Memory (4-bit): 4.4 GB → 20 GBCheaperGPT OSS 120BhighOpen+56%-14%$0.092Memory (4-bit): 4.4 GB → 70 GBModelScoreCostPer taskWhat changes
Qwen3.5 4BthinkingOpen+77%-14%$0.03 / $0.15Nemotron 3.5 LightningthinkingOpen+73%-30%$0.079Memory (4-bit): 4.4 GB → 20 GBGPT OSS 120BhighOpen+56%-14%$0.092Memory (4-bit): 4.4 GB → 70 GBGPT OSS 20BlowOpen+34%-49%$0.018 / $0.09Memory (4-bit): 4.4 GB → 13 GB
Granite 4.2 3BthinkingOpen+22%-25%$0.006Context: 131K → 128KModelScoreCostPer taskWhat changes
GPT-6 Lunalow×2.9+186%$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow×2.8+543%$0.01
Mistral Small 4thinkingOpen+52%+275%$0.015Memory (4-bit): 4.4 GB → 72 GB
MiMo v2.5thinkingOpen×2.9+150%$0.019Memory (4-bit): 4.4 GB → 186 GB
DeepSeek V3 0324non-thinkingOpen+31%+479%$0.021Context: 131K → 128K+1ModelScoreCostPer taskWhat changes
IK2 Horizon 7BthinkingOpen×2.8+650%$0.30 / $1.20Memory (4-bit): 4.4 GB → 4.5 GBIK2 Horizon 3.7BthinkingOpen×2.1+650%$0.30 / $1.20
Gemma 4 12BthinkingOpen+91%+114%$0.10 / $0.30Memory (4-bit): 4.4 GB → 6 GBSApriel v1.5 15B ThinkerOpen+86%——Context: 131K → 128K+1SApriel v1.6 15B ThinkerOpen+80%——Context: 131K → 128K+1ModelScoreCostPer taskWhat changes
Nemotron 3.5 Super VLOpen×3.6—$0.19Memory (4-bit): 4.4 GB → 75 GBNemotron 3 UltrathinkingOpen×3.1—$0.58Memory (4-bit): 4.4 GB → 336 GBNemotron 3.5 Bolt 74B-A9BOpen×2.9——Memory (4-bit): 4.4 GB → 45 GBNemotron 3.5 LightningthinkingOpen+73%—$0.079Memory (4-bit): 4.4 GB → 20 GBNemotron 3 Super 120B A12BOpen+73%—$1.64Memory (4-bit): 4.4 GB → 74 GBNemotron Cascade 2 30B A3BOpen+57%——Memory (4-bit): 4.4 GB → 19 GBNemotron 3 Nano Omni 30B A3B ReasoningOpen+38%—$0 / $0Memory (4-bit): 4.4 GB → 20 GBLlama Nemotron Super 49B v1.5non-thinkingOpensame—$0.40 / $0.40Context: 131K → 128K+1All NVIDIA models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · compressed✓ 16 GB · 8-bit✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
Licencenvidia-open-model-licensecommercial use with conditions
Benchmarks
all categoriesChat & Knowledge
Humanity's Last Exam4.9#282
AA Omniscience Index-41.6#179
CritPt (physics)0#180
AA Intelligence Index 4.3.2≈7.4#267
AA Omniscience · Accuracy11.8#269
AA-LCR (long context) 1.122.7#231
GPQA Diamond57#241
AIME 2025 · AA archived69.7#70
Coding
LiveCodeBench · AA archived72.4#38
Agents & Tool Use
τ²-bench · AA21.9#185
Writing & Language
IFBench (instructions)27.6#220
FAQ
How much does NVIDIA Nemotron Nano 9B V2 cost?›
$0.04 per 1M input tokens and $0.16 per 1M output tokens (cheapest host: DeepInfra).
Is NVIDIA Nemotron Nano 9B V2 free to use?›
Yes, there is a free API option.
How good is NVIDIA Nemotron Nano 9B V2, and what is it best at?›
#280 of 334 in Chat & Knowledge, #195 of 239 in Coding, #183 of 242 in Agents & Tool Use, #217 of 245 in Writing & Language.
Is NVIDIA Nemotron Nano 9B V2 better than Granite 4.1 30B?›
On AA Intelligence Index 4.3.2, NVIDIA Nemotron Nano 9B V2 scores 7.4 vs 7.4 — about the same.
How can I use NVIDIA Nemotron Nano 9B V2?›
Through the API of DeepInfra, NVIDIA.
What is the context window of NVIDIA Nemotron Nano 9B V2?›
131K tokens, with up to 131K output tokens.
When was NVIDIA Nemotron Nano 9B V2 released? Is there a newer version?›
Released 2025-08-18. It is the current version in its line.
Is NVIDIA Nemotron Nano 9B V2 open source?›
Yes, the weights are open (nvidia-open-model-license licence).
Can I run NVIDIA Nemotron Nano 9B V2 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can NVIDIA Nemotron Nano 9B V2 read images?›
No, it accepts text only.
What is NVIDIA Nemotron Nano 9B V2?›
NVIDIA Nemotron Nano 9B V2 is an open-weights language model from NVIDIA, released Aug 2025.
What are the best alternatives to NVIDIA Nemotron Nano 9B V2?›
Qwen3.5 4B (cheaper), GPT-6 Luna (better), K2 Horizon 7B (open weights).