nemotron-3.5-lightning:30b-a3b-mxfp8

943 5 hours ago

NVIDIA Nemotron 3.5 Lightning is an open 30B mixture-of-experts (MoE) model with 3B active parameters built for always-on agents.

tools thinking 30b 
ollama run nemotron-3.5-lightning:30b-a3b-mxfp8

Details

5 hours ago

906068bad076 · 34GB ·

template
{{ .Prompt }}
json
{ "architectures": [ "NemotronHForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "a
json
{ "_from_model_config": true, "bos_token_id": 1, "do_sample": true, "eos_token_id": [ 2, 11 ], "pad_
json
{ "bos_token": { "content": "<s>", "lstrip": false, "normalized": false, "rstrip": false, "single_wo
json
{ "version": "1.0", "truncation": null, "padding": null, "added_tokens": [ { "id": 0, "content": "<u
json
{ "add_bos_token": false, "add_eos_token": false, "add_prefix_space": false, "added_tokens_decoder":
license
NVIDIA Open Model Agreement TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION Release Dat
params
{ "temperature": 1, "top_p": 0.95 }
393 tensors

Readme

NVIDIA Nemotron 3.5 Lightning is an open 30B mixture-of-experts (MoE) model with 3B active parameters built for the execution layer of always-on agents. It is designed for harnesses like OpenClaw and Hermes Agent—all supported by the NVIDIA NemoClaw open source security and management stack for running always-on AI agents.

Nemotron 3.5 Lightning offers 4x higher throughput and 30% lower task completion time compared to other leading open models of similar size

Use cases

  • Personal agents: Power long-running personal assistants for everyday tasks, like managing your email, calendar, projects and bookings. Run these agents locally on your PC to leverage contextualized intel.
  • Financial services workflows: Power special agents for specialized financial services tasks, such as extracting data from documents, checking policy rules, monitoring risk signals, and preparing structured summaries.
  • Cybersecurity operations: Power special agents for specialized security tasks, such as enriching alerts, classifying incidents, querying logs, validating controls, correlating indicators, and preparing structured findings for analysts.
  • Telecom: Power special agents for autonomous networks and proactive customer care, such as triaging network alarms, optimizing network configurations, and answering billing questions.
  • Retail: Power special agents for enriching product catalogs, resolving inventory and fulfillment exceptions, assisting product discovery, and answering order, return, or loyalty questions

Benchmarks

image.png