lars

moedexDecision modelOpen weights

lars by moedex is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text.

Pricing

No hosted price is listed. The weights are open: the cost is your own compute.

Specifications

Provider: moedex
Parameters: —
Released: Sep 2026
Open weights: Yes
Question types: Choice (one of N options), Score (ordered rubric), Yes / no
Mechanism: Decoder LM with a decision readout
Backbone: Qwen/Qwen3-4B-Instruct-2507
Jev-compatible request: Yes
Runs on CPU: —
License: apache-2.0
Hugging Face: moedex/lars

Compare lars with…

lars vs amos-1-large—/1M in lars vs amos-1-medium—/1M in lars vs Celeris-1 Decision$0.04/1M in lars vs ScaleDown Classify-1$0.04/1M in lars vs Cloudflare clef—/1M in lars vs Cloudflare clef-flash—/1M in

FAQ

How much does lars cost?

lars has no hosted price listed; its weights are open, so the cost is the compute you run it on.

What questions can lars answer?

lars answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.

Is lars well calibrated?

No calibration measurement for lars is recorded yet.

Is lars open weight, and can it run on a CPU?

Yes, lars has open weights.

Data as of 2026-10-14. Source: Metatext decision-model catalogue, 2026-10-14 (Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.