Text Generation
Transformers
Safetensors
PyTorch
English
nemotron_labs_sigma
feature-extraction
nvidia
nemotron-labs
conversational
custom_code
Instructions to use nvidia/Nemotron-Labs-Sigma-3B-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use nvidia/Nemotron-Labs-Sigma-3B-instruct with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="nvidia/Nemotron-Labs-Sigma-3B-instruct", trust_remote_code=True) messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("nvidia/Nemotron-Labs-Sigma-3B-instruct", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use nvidia/Nemotron-Labs-Sigma-3B-instruct with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "nvidia/Nemotron-Labs-Sigma-3B-instruct" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nvidia/Nemotron-Labs-Sigma-3B-instruct", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/nvidia/Nemotron-Labs-Sigma-3B-instruct
- SGLang
How to use nvidia/Nemotron-Labs-Sigma-3B-instruct with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "nvidia/Nemotron-Labs-Sigma-3B-instruct" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nvidia/Nemotron-Labs-Sigma-3B-instruct", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "nvidia/Nemotron-Labs-Sigma-3B-instruct" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nvidia/Nemotron-Labs-Sigma-3B-instruct", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use nvidia/Nemotron-Labs-Sigma-3B-instruct with Docker Model Runner:
docker model run hf.co/nvidia/Nemotron-Labs-Sigma-3B-instruct
Commit History
2 days ago
-
Re-convert preserving fp32 geometry tensors; restore the SFT tokenizer on instruct
Verified signature057a084
6 days ago
-
Point evaluation results at the technical report; document the per-task prompting convention
Verified signature72abdaf -
Fix classifier-free guidance: isolate the unconditional branch, carry the raw conditional mean as self-conditioning, keep the readout unguided; ship the SFT tokenizer (chat control tokens, EOS <|im_end|>)
Verified signature7eb33c7 -
Cite the arXiv paper (2610.02665)
Verified signature786b7f0 -
Add the full author list and align the title with the technical report
Verified signature431e286 -
Set release date 11/12/2026
Verified signatured1fe0a9
7 days ago
-
Add self-speculative decoding to the shared inference code
Verified signaturee479524
8 days ago
-
Add recommended per-budget DDIM/CFG/ST settings
Verified signature49d6cf9
9 days ago
-
Fix Model Card++ subcards: correct model name, context length and NFE
Verified signaturef8233cc -
Model card: align repo references with renamed repos
Verified signature70e2e0f -
Fix max_position_embeddings 4096->8192 (SFT trained at seq 8192; YaRN factor 0.5)
Verified signatureb9c10d1 -
fix: declare projected_bernstein schedule parameterization
Verified signature5cd3aa4 -
MS-SUSReg lineage weights
Verified signature221782c
19 days ago
-
Remove old-name module file
Verified signature6923db9 -
Remove old-name module file
Verified signature3a81346 -
Rename family to Nemotron-Labs-Sigma (code identifiers, cards, configs)
Verified signatured28f3d7
25 days ago
-
private staging for legal review
Verified signature606a6b7 -
private staging for legal review
Verified signaturec8ca19e
26 days ago
-
private staging for legal review
Verified signature028a99c -
private staging for legal review
Verified signatureac45e35
27 days ago
-
private staging for legal review
Verified signaturec664a74
29 days ago
-
private staging for legal review
Verified signature3a05083
30 days ago
-
private staging for legal review
Verified signature4d39364
-
private staging for legal review
Verified signaturefd9676f -
initial commit
Verified signature255c66f