edtorre/lightning-agent

edtorre/ lightning-agent:latest

2 Downloads Updated yesterday

Lightning Agent — a 32.9B Nemotron MoE model tuned for Hermes agentic workflows.

ollama run edtorre/lightning-agent

curl http://localhost:11434/api/chat \
  -d '{
    "model": "edtorre/lightning-agent",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='edtorre/lightning-agent',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'edtorre/lightning-agent',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated yesterday

yesterday

86851e730642 · 25GB ·

model

archnemotron_h_moe

parameters32.9B

quantizationQ4_K_M

25GB

template

13B

license

NVIDIA Open Model License Agreement Last Modified: October 24, 2025 This NVIDIA Open Model License A

10kB

system

You are Hermes, a highly capable AI assistant and agent. You help users accomplish tasks by reasonin

306B

params

{ "draft_num_predict": 2, "num_ctx": 65536, "num_predict": -1, "repeat_penalty": 1.0

109B

Readme

Built from nemotron-3.5-lightning:latest with a 65,536-token context window, this model is configured for use as an agent (“Hermes”) with step-by-step reasoning, tool calling, and
chain-of-thought thinking capabilities. Sampling parameters are tuned for reliable, deterministic agent behavior: temperature 0.7, top-p 0.8, repeat penalty 1.05.

Capabilities: completion · tools · thinking

Use cases: autonomous agents, multi-step task execution, function-calling pipelines, reasoning-intensive workflows.

Built from nemotron-3.5-lightning:latest with a 65,536-token context window, this model is configured for use as an agent ("Hermes") with step-by-step reasoning, tool calling, and          
 chain-of-thought thinking capabilities. Sampling parameters are tuned for reliable, deterministic agent behavior: temperature 0.7, top-p 0.8, repeat penalty 1.05.                           
                                                                                                                                                                                              
 Capabilities: completion · tools · thinking                                                                                                                                                  
                                                                                                                                                                                              
 Use cases: autonomous agents, multi-step task execution, function-calling pipelines, reasoning-intensive workflows.

![Lightning_Agent.jpeg](/assets/edtorre/lightning-agent/3f04b8a2-22f9-45bc-b22f-3ff82a1fda99)

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)