[ad_1]
When we launched Llama-3.1-FoundationAI-SecurityLLM-base-8B (Foundation-sec-8B) in April, we proved that an eight-billion-parameter mannequin—educated solely on safety knowledge—can outperform general-purpose LLMs many occasions its measurement on cybersecurity benchmarks. Practitioners beloved the outcomes, but they saved asking:
“Can I interact with it like ChatGPT—no extra scaffolding, just prompt and go?”
Now you’ll be able to. Llama-3.1-FoundationAI-SecurityLLM-instruct-8B (Foundation-sec-8B-Instruct) layers instruction fine-tuning on prime of our domain-focused base mannequin, providing you with a chat-native copilot that understands safety context and follows natural-language instructions straight out of the field.
Built for Security, Tuned for Conversation
- Instruction following — Foundation-sec-8B-Instruct obeys on a regular basis prompts for summarization, question-answering, sentiment evaluation, and inventive textual content era with none additional fine-tuning.
- Role consciousness — It understands system, consumer, and assistant roles, enabling multi-turn chats, retrieval-augmented era, and agent-style purposes.
- Compact footprint — A 4K-token context window lets Foundation-sec-8B-Instruct run on a single, high-memory GPU whereas we push towards for much longer home windows.
- Open-source and light-weight — Like its sibling, Foundation-sec-8B-Instruct ships as absolutely open weights underneath a permissive license—sufficiently small for on-prem, air-gapped, or edge deployments with out vendor lock-in.
Performance That Outpaces Larger Models
Foundation-sec-8B-Instruct delivers heavyweight safety efficiency in an eight-billion-parameter type issue. In unbiased evaluations, it outperforms bigger fashions whereas remaining sufficiently small to run on a single high-memory GPU.
| Benchmark | Foundation-sec-8B-Instruct | Llama-3.1-8B-Instruct | Llama-3.1-70B-Instruct | GPT-4o-mini |
|---|---|---|---|---|
| CTI-RCM | 0.692 | 0.558 | 0.623 | 0.655 |
Just as importantly, Foundation-sec-8B-Instruct maintains sturdy general-language efficiency on broad assessments corresponding to MMLU. That means it could fluidly clarify menace conduct, help long-form investigation narratives, and reply naturally in chat workflows—with out compromising its safety focus or demanding outsized infrastructure.
Where It Fits Today
Security Operations Centers (SOC) face relentless alert noise, fragmented tooling, and power staffing gaps. In pilots throughout Cisco CSIRT and Cisco XDR, Foundation-sec-8B-Instruct classifies alerts, maps observables to MITRE ways, reconstructs timelines, and drafts investigation reviews—reducing false positives and accelerating triage so analysts can deal with actual threats.
Application Security (AppSec) groups juggle code critiques, menace fashions, and red-team planning with restricted skilled bandwidth. Early deployments at SBG Product Security, Meraki, and the S&TO Web-Application Security group use Foundation-sec-8B-Instruct to simulate assault paths, generate threat-model diagrams, overview code towards OWASP pointers, and craft customized payloads, shifting work from reactive fixes to proactive design.
Because Foundation-sec-8B-Instruct is instruction-tuned, most workflows want solely a well-crafted immediate or a easy retrieval template—no further coaching loop required.
Looking Ahead
Foundation-sec-8B-Instruct is a milestone, however it’s removed from the ultimate vacation spot. Our roadmap focuses on unlocking richer context, tighter construction, and broader knowledge modalities, so the mannequin will be capable to deal with each artifact defenders depend on—with out bolting on additional infrastructure. In the close to time period you’ll be able to count on:
- Longer context home windows — 16K+ tokens for full playbooks, SBOMs, and log bundles
- Structured interplay — Schema-validated JSON, express perform calling, and even code or binary era
- Multimodal inputs — Logs, packet captures, screenshots, and different artifacts in a single dialog
- Next-generation fashions — A reasoning-optimized eight billion-parameter variant and a forthcoming seventy billion-parameter mannequin
Start Building Today
- Grab the weights — Download Foundation-sec-8B-Instruct on Hugging Face and run it on-prem, in safe cloud enclaves, or in air-gapped labs—no licensing constraints.
- Follow the recipes — The Foundation AI Cookbook presents deployment guides, retrieval templates, and agent examples.
- Join the neighborhood — Pilot new workflows, contribute prompts or fine-tunes, and feed your findings again into the open-source ecosystem.
Foundation-sec-8B-Instruct is reside, open, and able to defend. Download it, immediate it, and assist form the way forward for AI-powered cybersecurity.
Contributors: Sajana Weerawardhena, Paul Kassianik, Blaine Nelson, Baturay Saglam, Anu Vellore, Aman Priyanshu, Supriti Vijay, Massimo Aufiero, Fraser Burch, Arthur Goldblatt, Konstantin Goldin, Alie Fordyce, Dhruv Kedia, Zhouran Yang, Ed Li, Jianliang He, Kojin Oshiba, Yaron Singer, Amin Karbasi
We’d love to listen to what you assume! Ask a query and keep linked with Cisco Security on social media.
Cisco Security Social Media
Share: