Nemotron 3.5 Lightning
Nemotron 3.5 LightningRankingin thinking mode?
Alternatives?
CheaperGPT-6 Lunamax×3-15%$0.068Function calling only with none · Chat Completions+3
OpenQwen3.8 27Bxhigh×2.6+1,167%$1.01Context: 1M → 262K+5
CheaperMiMo v2.6 FlashthinkingOpen×2.9-22%$0.062Memory (4-bit): 20 GB → 186 GB+4
CheaperClaude Haiku 5.5medium×2.7-41%$0.047Remove sampling parameters+5CheaperGPT-5.6 Lunahigh×2.5-45%$0.044GPT-6 Lunamax×3-15%$0.068Function calling only with none · Chat Completions+3MiMo v2.6 FlashthinkingOpen×2.9-22%$0.062Memory (4-bit): 20 GB → 186 GB+4Claude Haiku 5.5medium×2.7-41%$0.047Remove sampling parameters+5GPT-5.6 Lunahigh×2.5-45%$0.044MiMo v2.5 ProthinkingOpen×2-32%$0.054Memory (4-bit): 20 GB → 614 GBGPT-6 Lunalow+67%-94%$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow+63%-88%$0.01MiMo v2.5thinkingOpen+70%-76%$0.019Memory (4-bit): 20 GB → 186 GBClaude Haiku 5.5low×2.3-69%$0.024Remove sampling parameters+5GPT-5 Minihigh+30%-33%$0.053Context: 1M → 400KQwen3.8 27BxhighOpen×2.6+1,167%$1.01Context: 1M → 262K+5IK2 Horizon MoVA 36B A4BthinkingOpen+97%+3,490%$1 / $4Context: 1M → 524K+1Qwen 3.5 27BthinkingOpen+78%+1,592%$0.30 / $2.40Context: 1M → 262KAG9v3 39A5BOpen+70%——Context: 1M → 131K+1Qwen3.6 27BthinkingOpen+67%+682%$0.62Context: 1M → 262KNemotron 3.5 Super VLOpen×2.1+137%$0.19Memory (4-bit): 20 GB → 75 GBNemotron 3 UltrathinkingOpen+78%+626%$0.58Memory (4-bit): 20 GB → 336 GBNemotron 3.5 Bolt 74B-A9BOpen+66%——Context: 1M → 256K+1Nemotron 3 Super 120B A12BOpensame+1,962%$1.64Memory (4-bit): 20 GB → 74 GBNemotron Cascade 2 30B A3BOpen-9%——Nemotron 3 Nano Omni 30B A3B ReasoningOpen-20%—$0 / $0Context: 1M → 256KLlama Nemotron Super 49B v1.5non-thinkingOpen-43%—$0.40 / $0.40Context: 1M → 128K+1Llama 3.1 Nemotron Nano 4B v1.1Open-43%——Context: 1M → 128KAll NVIDIA models →Run it yourself
estimateBenchmarks
all categoriesFAQ
How much does Nemotron 3.5 Lightning cost?›
$0.025 per 1M input tokens and $0.12 per 1M output tokens (cheapest host: Darkbloom). A typical benchmark task costs about $0.079 at default effort.
Is Nemotron 3.5 Lightning free to use?›
There is no free API; the cheapest paid price is $0.025 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Nemotron 3.5 Lightning, and what is it best at?›
#182 of 334 in Chat & Knowledge, #63 of 132 in Coding, #113 of 133 in Agents & Tool Use, #101 of 133 in Vision & Documents.
Is Nemotron 3.5 Lightning better than GPT-5 Nano?›
On AA Intelligence Index 4.3.2, Nemotron 3.5 Lightning scores 12.9 vs 13 — about the same.
How can I use Nemotron 3.5 Lightning?›
Through the API of CoreWeave, Darkbloom, DeepInfra.
What is the context window of Nemotron 3.5 Lightning?›
1M tokens.
When was Nemotron 3.5 Lightning released? Is there a newer version?›
Released 2026-08-11. It is the current version in its line.
Is Nemotron 3.5 Lightning open source?›
Yes, the weights are open (openmdw-1.1 licence).
Can I run Nemotron 3.5 Lightning locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Nemotron 3.5 Lightning read images?›
No, it accepts text only.
What is Nemotron 3.5 Lightning?›
Nemotron 3.5 Lightning is an open-weights language model from NVIDIA, released Aug 2026.
What are the best alternatives to Nemotron 3.5 Lightning?›
GPT-6 Luna (cheaper), Qwen3.8 27B (open weights).