# SAGEA: mira

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

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

## Facts

| | |
|---|---|
| Maker | SAGEA (https://systemonemodels.tech/sagea) |
| Decides | choice, score, noul, classify, route |
| Architecture | mira |
| Base model | jhu-clsp/mmbert-small |
| Licence | apache-2.0 |
| Availability | Open weights |
| Latest version | 0.1.4 |

## Reported evaluation

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

- Decision accuracy: 74.5%
- Calibration error (ECE): 0.025
- Median latency: 30 ms
- p95 latency: 43 ms

## Model card

# Mira-v1

**Fast, honest, open typed decisions**

## What is this?

Mira turns **state + typed questions → probability distributions** in a single
forward pass. No text generation. Ask `choice`, `score`, or `noul` questions —
alone or batched — and get calibrated probabilities, confidence, and an
abstain signal your code can branch on.

## Evaluation

Measured on commodity CPU (i5-1335U), one harness for every open model.

| Benchmark | Mira-v1 | Julia 1 | Jev  |
|---|---:|---:|---:|
| Typed decisions (2000) | **0.745** | 0.726 | 0.727 |
| ECE (15-bin) | **0.025** | 0.234 | — |
| Confident-error (≥0.9) | **0.006** | 0.234 | — |
| AG News (100) | 0.25 | 0.94 | 0.91 |
| Emotion (100) | **0.87** | 0.86 | 0.48 |
| Banking77 (100, native full-label) | **0.93** | 0.64 (72-label shortlist variant) | 0.87 |
| Short-query p50 | **30 ms** | 321 ms | 70–500 ms API |


romanized Nepali, and code-mixed slices are Mira-first reporting no other
small model publishes.

## Limitations

Mira compares answers you provide. it cannot supply missing facts or do
arithmetic. Severity direction and urgency separation are the softest slices.
Past 20 options a two-stage shortlist narrows then decides (recall reported).
Score the exact questions you plan to use before consequential actions.
Training data is private; eval harness and predictions are public.

---

---

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