- Size
- 31.6B parameters, 3.0B active (from the repo name)counted from the safetensors header on the Hub
- Licence
- custom, read it, OpenMDW License Agreement, version 1.1Our reading, check the licence: A custom licence we have not read for this model: other makers use the same name for different texts. Read the model's own licence file before you use it.Licence text
- On the Hub since
- 1 Aug 2026the date the repo was created
- Context length
- 256k tokensfrom the model's config
- Kind
- Base model (no base model declared on the Hub)NemotronHForCausalLM
- On Hugging Face
- nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16941k downloads in the last 30 days, 279 likes
Scores
No independent scores for this model in Epoch AI's data. A score is only shown here when Epoch's name for a model matches this repo with certainty; an uncertain match is left out rather than guessed.
Claimed by the maker Claimed
The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.
Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.
Run it yourself
Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 0.41 GB, worked out from the model's config (nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16), at 16-bit precision. Choose a machine above to see a verdict for each format. Method
GGUF for llama.cpp, Ollama, LM Studio and others
mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF (converted by mradermacher), 2.7k downloads. Sizes are the real files on the Hub, all parts added up.
| Format | Size | Memory needed | Get it |
|---|---|---|---|
| Q2_K | 17.4 GBfile size | 18.3 GB17.4 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q2_KOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q2_KDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q2_K.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q3_K_S | 17.4 GBfile size | 18.3 GB17.4 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q3_K_SOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q3_K_SDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q3_K_S.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| IQ4_XS | 17.7 GBfile size | 18.6 GB17.7 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:IQ4_XSOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:IQ4_XSDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.IQ4_XS.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q3_K_M | 19.3 GBfile size | 20.2 GB19.3 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q3_K_MOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q3_K_MDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q3_K_M.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q3_K_L | 20.1 GBfile size | 21.0 GB20.1 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q3_K_LOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q3_K_LDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q3_K_L.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q4_K_S | 21.3 GBfile size | 22.2 GB21.3 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q4_K_SOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q4_K_SDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q4_K_S.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q5_K_S | 23.1 GBfile size | 24.0 GB23.1 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q5_K_SOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q5_K_SDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q5_K_S.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q4_K_M | 23.7 GBfile size | 24.6 GB23.7 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q4_K_MOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q4_K_MDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q4_K_M.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q5_K_M | 25.2 GBfile size | 26.1 GB25.2 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q5_K_MOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q5_K_MDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q5_K_M.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q6_K | 32.5 GBfile size | 33.4 GB32.5 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q6_KOllama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q6_KDownload hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q6_K.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
| Q8_0 | 32.6 GBfile size | 33.5 GB32.6 + 0.41 + 0.50 | Commandsllama.cpp llama-server -hf mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q8_0Ollama ollama run hf.co/mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF:Q8_0Download hf download mradermacher/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF --include "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16.Q8_0.gguf" --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-GGUF |
MLX for Apple silicon
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| MLX 4bitKagandi/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-mlx-4Bit | 16.6 GBfile size | 17.5 GB16.6 + 0.41 + 0.50 | Download page |
| MLX 8bitMcG-221/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16-mlx-8Bit | 31.3 GBfile size | 32.2 GB31.3 + 0.41 + 0.50 | Download page |
Original weights size from the safetensors header, for vLLM, SGLang, Transformers
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| BF16original weights | 61.3 GBfile size | 62.2 GB61.3 + 0.41 + 0.50 |
hf download nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 --local-dir NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16The Hub tags 100 or more quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.
Run it by API
4 providers serve this model through the Hugging Face router. Prices are US dollars per million tokens. Precision is what the provider reports; a lower precision than the original weights can change answers. Live first, then cheapest output.
| Provider | In | Out | Context | Precision | Tokens/s | First token | Uptime 24h | Tools | JSON |
|---|---|---|---|---|---|---|---|---|---|
| DeepInfra | $0.06 | $0.16 | 256k | bf16 | 159 | 0.3 s | 99.9% | yes | yes |
| CoreWeave | $0.07 | $0.20 | 256k | bf16 | 105 | 0.5 s | 100.0% | yes | yes |
| Fireworks | $0.05 | $0.20 | 256k | not given | 194 | 0.3 s | 100.0% | yes | yes |
| ModelRun | $0.07 | $0.20 | 1M | fp4 | 153 | 0.2 s | 100.0% | yes | yes |
Through the router, with a Hugging Face token (the router bills your Hugging Face account). Change the name after the colon to pick another provider.
curl https://router.huggingface.co/v1/chat/completions \
-H "Authorization: Bearer $HF_TOKEN" -H "Content-Type: application/json" \
-d '{"model": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16:deepinfra", "messages": [{"role": "user", "content": "Hello"}]}'Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.
Planned for this page Not built yet
- Try it in private chat
- A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
- Fine-tune it
- Fine-tuning a few base models from this page and publishing the result here.
- Creators and earnings
- Pages for people who publish fine-tunes and adapters of this model, and what they earn.
- Fund and bounties
- Bounties for new open models, paid from a public fund. How the fund will work.
- Contribute compute
- How to give compute to open model work, with Akash.
Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.