DedeProGames/Astral-2.6:3b

DedeProGames/ Astral-2.6:3b

20 Downloads Updated 4 days ago

Astral-2.6 is a 3B-parameter model built for high performance on general reasoning tasks, with a special focus on long and chronological reasoning

tools thinking 3b

ollama run DedeProGames/Astral-2.6:3b

curl http://localhost:11434/api/chat \
  -d '{
    "model": "DedeProGames/Astral-2.6:3b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='DedeProGames/Astral-2.6:3b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'DedeProGames/Astral-2.6:3b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 4 days ago

4 days ago

478ec94ab06b · 4.2GB ·

model

archllama

parameters3.93B

quantizationQ8_0

4.2GB

template

{{- $lastUserIdx := -1 -}} {{- range $idx, $msg := .Messages -}} {{- if eq $msg.Role "user" }}{{ $la

2.1kB

license

Could not load data.

params

{ "min_p": 0.01, "num_ctx": 65536, "repeat_penalty": 1, "stop": [ "<|im_star

148B

Readme

Astral-2.6 is a 3B-parameter model built for high performance on general reasoning tasks, with a special focus on long, chronological reasoning (multi-step timelines, event ordering, long-range dependencies, and consistency over extended contexts). It is engineered to deliver advanced capability for its size, while remaining practical on smaller devices and modest hardware.

Strong general reasoning: robust across logic, problem solving, and structured analysis.
Long chronological reasoning: designed to handle multi-stage temporal chains (before/after constraints, timelines, causality, state tracking).
Efficient by design: tuned for excellent quality-per-parameter, enabling strong results without requiring large GPUs.
Small-hardware friendly: intended to run well on compact desktops, laptops, and edge devices that can host 3B-class models.