dgpl/ dgpl-linux-assistant:q6_k

82 1 week ago

DGPL Linux Assistant 0.8B is an ultra-fast, expert, and production-safe Linux terminal copilot developed by Durbhasi Gurukulam Private Limited (DGPL). Runs 100% locally on standard CPU and RAM (512MB-1GB RAM).

vision 0.8b 
ollama run dgpl/dgpl-linux-assistant:q6_k

Details

1 week ago

02ca39ff436d · 835MB

model
qwen35
·
752M
·
Q6_K
projector
clip
·
101M
·
F16
template
{{- range .Messages }} {{- if eq .Role "system" }}<|im_start|>system {{ .Content }}<|im_end|> {{- el
system
You are DGPL Linux Assistant 0.8B, an ultra-fast Linux Terminal Copilot and DevSecOps systems assist
params
{ "num_predict": 768, "repeat_last_n": 64, "repeat_penalty": 1.18, "stop": [

Readme

DGPL Linux Assistant 0.8B

DGPL Linux Assistant 0.8B is an ultra-fast, lightweight, and production-safe Linux terminal copilot developed by Durbhasi Gurukulam Private Limited (DGPL). Built specifically for DevOps engineers, Red Teamers, sysadmins, and Linux developers, it delivers accurate, clean command-line actions and deep system diagnostics with zero bloat.

Choosing the Right Model for Your Hardware

Select the optimal model tag based on your available system memory (RAM / CPU):

Hardware / Resource Tier Recommended Tag Size RAM Required Best Use Case
Ultra-Lightweight / Edge (Default) dgpl/dgpl-linux-assistant:q4_k_m (or default) 505 MB 512 MB - 1.0 GB Budget VPS, standard laptops, CI/CD runners, and edge devices.
Balanced / Daily Workstation dgpl/dgpl-linux-assistant:q5_k_m 578 MB 1.0 GB - 1.2 GB Ideal balance of speed and complex regex / kernel reasoning.
High Precision Near-Lossless dgpl/dgpl-linux-assistant:q6_k 629 MB 1.2 GB - 1.5 GB High-precision output for complex sysadmin & network debugging.
Near-FP16 Accuracy dgpl/dgpl-linux-assistant:q8_0 775 MB 1.5 GB - 2.0 GB Near-FP16 fidelity with negligible quantization loss on CPU.
Full Precision Master Baseline dgpl/dgpl-linux-assistant:f16 1.51 GB 2.0 GB - 3.0 GB Uncompressed master 16-bit baseline weights for maximum precision.

Quickstart Commands by Tier

# 1. Default Ultra-Fast Lightweight (Q4_K_M - ~505 MB)
ollama run dgpl/dgpl-linux-assistant

# 2. Balanced Medium Quantization (Q5_K_M - ~578 MB)
ollama run dgpl/dgpl-linux-assistant:q5_k_m

# 3. High Precision Near-Lossless (Q6_K - ~629 MB)
ollama run dgpl/dgpl-linux-assistant:q6_k

# 4. High Fidelity Near-FP16 (Q8_0 - ~775 MB)
ollama run dgpl/dgpl-linux-assistant:q8_0

# 5. Full Precision 16-Bit Master Baseline (F16 - ~1.51 GB)
ollama run dgpl/dgpl-linux-assistant:f16

Key Capabilities

  • Command Generation & Explanation: Instant Linux CLI flags, grep, awk, find, sed, systemctl, journalctl, and raw network/socket syntax.
  • Offensive Security & Red Teaming: Terminal reconnaissance (nmap, netstat, ss, ip), POSIX permissions audit, and process monitoring.
  • DevOps & Cloud-Native Automation: Docker, Kubernetes (kubectl), systemd unit configuration, disk space (df, du), and log parsing.
  • Ultra-Fast & CPU Friendly: Engineered to run smoothly on edge servers, low-spec cloud VMs, and standard laptops. —

Python & API Usage

import ollama

response = ollama.chat(
    model="dgpl/dgpl-linux-assistant",
    messages=[
        {"role": "user", "content": "How do I check open listening ports with PID in Linux?"}
    ]
)

print(response["message"]["content"])

Corporate Entity & Support

  • Developed By: Durbhasi Gurukulam Private Limited (DGPL)
  • Official Support Email: support@durbhasigurukulam.com
  • Official Contact Phone: +91 7852034945
  • Headquarters: Jaipur, Rajasthan, Bharat (India)