# Speko 5 Hacker News commenters recommended Speko and 2 criticized it, November 2025 to November 2026. It was recommended in 1 of 13 months. **Verdict:** Benchmarks voice AI models continuously and routes each request to whichever performs best for your language and constraints. Critics question whether the value holds beyond the initial model choice. ## Why people recommend it - Continuous model benchmarks drive routing (3 comments): "we benchmark the models continuously and route based on those measurements for your language and constraints" ([source](https://news.ycombinator.com/item?id=49333622)) "The evals are the value, agreed. The routing is what makes them actionable" ([source](https://news.ycombinator.com/item?id=49333994)) "route each request to whatever wins for your language and constraints" ([source](https://news.ycombinator.com/item?id=49350874)) - Interesting concept, looks promising (3 comments): "Looks awesome, can't wait to try it for some Filipino workflows when it's available!" ([source](https://news.ycombinator.com/item?id=49333760)) "the concept is interesting I must say" ([source](https://news.ycombinator.com/item?id=49333353)) "This looks really interesting." ([source](https://news.ycombinator.com/item?id=49340337)) ## What people warn about - Questionable value after initial model choice (1 comments): "So something like OpenRouter for voices offer questionable value given that stakeholders usually make this sort of decisions once at the start of the project." ([source](https://news.ycombinator.com/item?id=49333588)) ## Alternatives hackers recommend - [Unpod.ai](https://hackerapproved.com/p/unpod-ai.md): 8 recommend - [Asterisk AI Voice Agent](https://hackerapproved.com/p/asterisk-ai-voice-agent.md): 7 recommend - [Dograh](https://hackerapproved.com/p/dograh.md): 6 recommend - [Vapi](https://hackerapproved.com/p/vapi.md): 6 recommend ## Cite as Hacker Approved, "Speko", https://hackerapproved.com/p/speko/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).