So, given that laya is not at all an LLM, we made it in a separate allocation: https://laya.blablador.fz-juelich.de

Given that this is a popular, and fast way of running decisions, I am considering other jev-like models. If you have suggestions, let me know!


Alex


On 23. Sep 2026, at 22:44, Strube, Alexandre <a.strube@fz-juelich.de> wrote:

Jev[1] took the AI world by storm lately, even though it’s not necessarily the 
WOW people are making out of it. It can be extremely useful, though.

In any case, 2 days after Jev came out, someone had an open-source equivalent of it: Laya [2].

And now, you can try laya too. It’s alias-laya on the api, the endpoint being identical to Jev.

Here is an example:

curl https://api.blablador.fz-juelich.de/v1/systemone \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "alias-laya",
    "state": "Write an extremely ultra advanced script optimizing multi-threaded CUDA kernels for matrix multiplication.",
    "questions": {
      "route_selection": {
        "type": "choice",
        "instructions": "Which capability tier is required to successfully fulfill this request?",
        "criteria": {
          "fast_tier": "Simple coding scripts, basic factual lookups, and clear formatting tasks.",
          "heavy_tier": "Highly complex logic, multi-threaded optimization, math proofs, programming, or ambiguous system architecture queries."
        }
      }
    }
  }' | jq

And the reply will be something like 

  "model": "laya-rl-agent",
  "answers": {
    "route_selection": {
      "type": "choice",
      "choice": "heavy_tier",
      "probabilities": {
        "fast_tier": 0.4083,
        "heavy_tier": 0.5917
      },
      "confidence": 0.0244,
      "action": {
        "act_probability": 1.0
      }
    }
  },
  "usage": {
    "input_tokens": 82,
    "output_tokens": 0
  },
  "routing": {
    "model": "english",
    "repo": "convaiinnovations/laya",
    "reason": "English Latin text",
    "detection": {
      "script": "latin",
      "script_profile": {
        "latin": 1.0
      },
      "language": "en",
      "is_english": true,
      "language_undecided": false,
      "diacritic_rate": 0.0,
      "non_latin_fraction": 0.0
    },
    "workflow": null
  }
}

And this query will run in… I don’t know, 0.05 seconds or so. 

Hope it’s useful! Have fun.




Let’s bark!

Alex, Blablador honcho