Loading comparison…
HEAD TO HEAD
Maisa: djev and SAGEA: mira, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | maisa/djev | sagea/mira |
|---|---|---|
| Summary | A typed-decision layer for Google's DiffusionGemma, from David Villalón at Maisa AI. It compiles a request into a small answer canvas, runs one denoising read on patched vLLM and reads the probabilities of the allowed labels, for text, images and images offered as options. | Small open typed-decision model with calibrated probabilities, first-class Nepali coverage. |
| Decides | choice, score, noul | choice, score, noul, classify, route |
| Architecture | djev | mira |
| Fine-tuned from | google/diffusiongemma-26b-a4b-it | jhu-clsp/mmbert-small |
| License | apache-2.0 | apache-2.0 |
| Availability | Open weights + hosted API | Open weights |
| Hosted by | Maisa | — |
| Input price | $0.035/MTok | — |
| Decision accuracy | — | 74.5% |
| Calibration error | — | 0.025 |
| Valid action rate | — | 100.0% |
| Median latency | — | 30 ms |
| p95 latency | — | 43 ms |
| Evaluation suite | — | s1-decision-bench |
| Latest version | 2026.09 | 0.1.4 |
Figures are from each model’s manifest; accuracy and latency are what the publishers report, on their own suites and hardware. Add a third model.
djev is from Maisa and mira from SAGEA. djev has open weights and a hosted API; mira has open weights you can download and run. Both answer choice, score and noul questions. Only mira answers classify and route.
Only mira publishes an accuracy figure (74.5% on s1-decision-bench); djev does not, so there is no comparison to make without your own test.
djev: $0.035 / $0 per 1M, or free to self-host. mira: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull maisa/djev and systemone pull sagea/mira download the weights.
| Variants | — | — |
|---|
| Size of latest version | — | 568.9 MB |
|---|
| Files | 0 | 13 |
|---|
| Downloads | 0 | 7 |
|---|
| Stars | 0 | 1 |
|---|
| Tags | system-one, diffusion, gemma, vllm, multimodal, training-free, 26b | routing, nepali, nepal, calibration, cpu |
|---|
| Updated | Sep 30, 2026 | Oct 1, 2026 |
|---|