# SAGEA: mira

> Small open typed-decision model with calibrated probabilities, first-class Nepali coverage.

- Page: https://systemonemodels.ai/sagea/mira
- API: https://api.systemonemodels.ai/v1/models/sagea/mira
- Download: `pip install systemonemodels && systemone pull sagea/mira`

## Facts

| | |
|---|---|
| Maker | SAGEA (https://systemonemodels.ai/sagea) |
| Decides | choice, score, noul, classify, route |
| Architecture | mira |
| Base model | jhu-clsp/mmbert-small |
| Licence | apache-2.0 |
| Availability | Open weights |
| Latest version | 0.1.5 |

## Reported evaluation

Suite: s1-decision-bench. Numbers are the publisher's own.

- Decision accuracy: 61.3%
- Calibration error (ECE): 0.083
- Median latency: 30 ms
- p95 latency: 43 ms

## Model card

# Mira v12 (System One package 0.1.5)

Small open typed-decision model (140M): choice / score / noul with calibrated
probabilities, CPU-first, Apache-2.0. Hugging Face: `sagea-ai/Mira-v1`.

## Measured (2026-10-02, box-01 CPU, native code)

- Neutral (typed-decisions rev c76749ec, 2000 Q): acc **0.6125** [0.589, 0.634],
  ECE 0.083, Brier 0.495 — choice 0.688 / score 0.618 / noul 0.530.
- Pilots: AGNews 0.23, Emotion 0.87, Banking77 0.93.
- Latency: same architecture as 0.1.4 (per-option cross-encoder).

## Limits (honest)

Neutral accuracy trails Julia-1 (0.726) and Laya (0.759). Strengths are
calibration (ECE 0.083 vs 0.2+ elsewhere), CPU speed, size, and support-style
routing (grid 0.745 lineage). Conventions (v11+): choice encodes
"name: description"; score levels read anchored ("level i of n-1: ...").

Run: `from mira_infer import MiraDecider; MiraDecider(...).decide(state, questions)`.

---

From System One Models — https://systemonemodels.ai/ · every System One model: https://systemonemodels.ai/system-one-models