Framer AI AgentsDesign and publish professional sites with AI
Promoted
Maker
📌
Hey Product Hunt!
Im the founder behind Zenic, we've been building Azno, its our own LLM line, and today were launching Azno 1.5 mini, LLM focued on coding\general purpose use (plus a little personality baked in, since we CARE how the model actually feels talk to, not just like at benchmarks!)
A few things that make this one different:
1. Built by a small team that also runs our own Trust & Safety stack, we spent a lot of time reducing false positives on totally bengin requests instead of just cranking up refusals across the board
2. We're still iterating hard on personality\roleplay behavior, so early feedback here is genuinely shapes the next update.
Would love to hear what you think would make this useful for your workflow, whats missing.
We're reading everything today!
Report
Have you considered adding a community prompt-sharing board so users can remix and upvote the best Azno model prompts? Would make the platform feel more lived-in and help newcomers learn faster from real examples.
Report
A built-in prompt history sidebar would be really helpful, something where I can scroll back through my past sessions, tag them, and pick up where I left off without losing context.
Report
the in-house safety evals piece is honestly refreshing, like most teams just hand-wave that part. nice to see someone actually building the guardrails themselves instead of tacking them on after
Report
honestly the responses feel surprisingly quick and the safety evals thing is a nice touch, gives me more confidence when throwing weird prompts at it
Report
The Azno models feel surprisingly snappy for something running in-house, and the built-in safety evals are a nice touch instead of being a separate add-on.
Have you considered adding a community prompt-sharing board so users can remix and upvote the best Azno model prompts? Would make the platform feel more lived-in and help newcomers learn faster from real examples.
A built-in prompt history sidebar would be really helpful, something where I can scroll back through my past sessions, tag them, and pick up where I left off without losing context.
the in-house safety evals piece is honestly refreshing, like most teams just hand-wave that part. nice to see someone actually building the guardrails themselves instead of tacking them on after
honestly the responses feel surprisingly quick and the safety evals thing is a nice touch, gives me more confidence when throwing weird prompts at it
The Azno models feel surprisingly snappy for something running in-house, and the built-in safety evals are a nice touch instead of being a separate add-on.