HEAD TO HEAD
Yuichi Tateno (hotchpotch): bekko-system-one and Convai Innovations: laya, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | hotchpotch/bekko-system-one | convai-innovations/laya |
|---|---|---|
| Summary | Ultra-small English encoders (17M, 68M and 395M) fine-tuned from Ettin rerankers. Each answers choice, noul and score questions with a probability per candidate and no generated text. Experimental v0 by one developer; the weights have no licence assigned yet. | An open-weights System One model from Convai Innovations. A fully fine-tuned ModernBERT-large encoder with a from-scratch decision head that scores one marker per option and answers every question in a single 33–39 ms pass. Runs on your own CPU or GPU. |
| Decides | choice, noul, score | choice, score, noul, classify, route |
| Architecture | bekko | laya |
| Fine-tuned from | cross-encoder/ettin-reranker-400m-v1 | answerdotai/modernbert-large |
| License | Unassigned: the weights licence is not finalised (treated as non-commercial); training code MIT | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | — | — |
| Calibration error | — | — |
| Valid action rate | — | — |
| Median latency | 5 ms | 39.5 ms |
| p95 latency | 5.4 ms | — |
| Evaluation suite | The maker's short-input speed probe (4-option Choice, RTX 5090, 1 Oct 2026) | — |
| Latest version | 2026.09 | 2026.09.25 |
| Variants | 0_BekkoInference | multilingual, tokenizer, eval, typed-decisions, encoder |
| Size of latest version | 1.5 GB | 2.2 GB |
| Files | 9 | 19 |
| Downloads | 0 | 431 |
| Stars | 0 | 0 |
| Tags | system-one, encoder, ettin, onnx, browser, small, 395m | system-one, encoder, modernbert, rlcd, multilingual, 421m |
| Updated | Oct 7, 2026 | Nov 22, 2026 |
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.
bekko-system-one is from Yuichi Tateno (hotchpotch) and laya from Convai Innovations. Both have open weights you can download and run. Both answer choice, noul and score questions. Only laya answers classify and route. bekko-system-one is the smaller model, at 395M parameters to 421M. bekko-system-one is licensed other; laya, apache-2.0.
Neither publishes an accuracy figure. Test both on your own labelled examples.
bekko-system-one: Free (open weights). laya: Free (open weights). Open weights cost nothing per call beyond your own hardware.
By their publishers’ figures, bekko-system-one answers in about 5 ms at the median and laya in about 39.5 ms — measured on different hardware, so treat it as a rough guide.
Yes, both: systemone pull hotchpotch/bekko-system-one and systemone pull convai-innovations/laya download the weights.