🧪 Explore the beta, open a PR and join the community.

Full announcement 👇

🚀 This is only v0.1. We’re releasing early because we believe in building in the open.

🛠️ Coming soon: jailbreak detection, toxicity scoring, and deeper integrations with Opik’s evaluation tools.

🛡️Meet Opik Guardrails (Public Beta) - a lightweight, open-source toolkit for real-time LLM output management:

⚠️Topic moderation via a fine-tuned BART model
🔎 PII detection built on Presidio
💪 Support for third-party guardrail libraries

Prompt engineering is evolving fast. We’re building tools that help you evolve with it 🔥

🛠️ Opik Agent Optimizer is open source — feedback, ideas, and PRs are all welcome.

Full announcement 👉https://www.comet.com/site/blog/automated-prompt-engineering/

Ships with 4 optimizers – including 3 developed in-house:

🧠 MIPRO – meta-prompt optimizer
🧪 Meta Prompter Optimizer – LLM works as a prompt engineer
🎯 Bayesian Few-Shot Optimizer – finds optimal few-shot examples
🧬 Evolutionary Optimizer

🤝 Meet Opik Agent Optimizer, now in public beta 🎉

A new SDK in Comet’s open-source Opik framework to automate prompt and agent tuning.

✅ Framework-agnostic
✅ Model-agnostic
✅ Plugs into any LLM stack
✅ Powered by Opik’s eval capabilities

Mirror lets teens:
📝Express themselves through writing, voice, images, or video
☀️ Track their moods
💡Get guided prompts to process feelings

Learn more 🔗

I used the LLM Jury to automatically evaluate my outputs using @comet.com's Opik, a free, 100% open-source LLM evaluation framework.

⭐️ Check it out and give it a star if you like what you see: (6/6)

github.com/comet-ml/opik

📅 March 6 | 1-2 PM EST
🔗 Register: aws-experience.com/amer/smb/e/8...