Loading comparison…
HEAD TO HEAD
Blockbrain Labs: cygnet and SAGEA: mira, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | blockbrain-labs/cygnet | sagea/mira |
|---|---|---|
| Summary | Typed decisions from Google's frozen Gemma-4-12B-it, with no fine-tuning. A small shim over unmodified vLLM letters the options, reads the model's own probability for each letter at one answer position and applies one calibration temperature; one output token per decision. | Small open typed-decision model with calibrated probabilities, first-class Nepali coverage. |
| Decides | choice, score, noul | choice, score, noul, classify, route |
| Architecture | cygnet | mira |
| Fine-tuned from | google/gemma-4-12b-it | jhu-clsp/mmbert-small |
| License | mit | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | 87.9% | 74.5% |
| Calibration error | — | 0.025 |
| Valid action rate | — | 100.0% |
| Median latency | 50 ms | 30 ms |
| p95 latency | — | 43 ms |
| Evaluation suite | JevBench public set (231 items), JevBench CLI | s1-decision-bench |
| Latest version | 2026.09 |
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.
cygnet is from Blockbrain Labs and mira from SAGEA. Both have open weights you can download and run. Both answer choice, score and noul questions. Only mira answers classify and route. cygnet is licensed mit; mira, apache-2.0.
They report on different suites — cygnet 87.9% on JevBench public set (231 items), JevBench CLI, mira 74.5% on s1-decision-bench — so the numbers do not rank them. Test both on your own labelled examples.
cygnet: Free (open weights). mira: Free (open weights). Open weights cost nothing per call beyond your own hardware.
By their publishers’ figures, mira answers in about 30 ms at the median and cygnet in about 50 ms — measured on different hardware, so treat it as a rough guide.
Yes, both: systemone pull blockbrain-labs/cygnet and systemone pull sagea/mira download the weights.
| 0.1.4 |
| Variants | — | — |
|---|
| Size of latest version | — | 568.9 MB |
|---|
| Files | 0 | 13 |
|---|
| Downloads | 0 | 7 |
|---|
| Stars | 0 | 1 |
|---|
| Tags | system-one, gemma, vllm, jevbench, training-free, 12b | routing, nepali, nepal, calibration, cpu |
|---|
| Updated | Sep 30, 2026 | Oct 1, 2026 |
|---|