HEAD TO HEAD
Yuichi Tateno (hotchpotch): bekko-system-one and Maziyar Panahi: modernjev-decide, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | hotchpotch/bekko-system-one | maziyar-panahi/modernjev-decide |
|---|---|---|
| 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. | A 149.6M ModernBERT-base cross-encoder with a scalar head that picks an agent's next action or tool from labels and descriptions you supply. Each option is scored against the state and a softmax ranks the set; no text is generated. Experimental preview, not calibrated. |
| Decides | choice, noul, score | choice, rank, route |
| Architecture | bekko | modernjev |
| Fine-tuned from | cross-encoder/ettin-reranker-400m-v1 | answerdotai/modernbert-base |
| 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 | — | 73.4% |
| Calibration error | — | — |
| Valid action rate | — | — |
| Median latency | 5 ms | — |
| p95 latency | 5.4 ms | — |
| Evaluation suite | The maker's short-input speed probe (4-option Choice, RTX 5090, 1 Oct 2026) | AgentToolDecisions-180K held-out test, next action type (1,158 decisions, 3 choices; the maker's split) |
| Latest version | 2026.09 | 2026.09 |
| Variants | 0_BekkoInference | — |
| Size of latest version | 1.5 GB | 574.1 MB |
| Files | 9 | 5 |
| Downloads | 0 | 0 |
| Stars | 0 | 0 |
| Tags | system-one, encoder, ettin, onnx, browser, small, 395m | system-one, encoder, modernbert, cross-encoder, tool-routing, preview, 149.6m |
| Updated | Oct 7, 2026 | Oct 7, 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 modernjev-decide from Maziyar Panahi. Both have open weights you can download and run. Both answer choice questions. Only bekko-system-one answers noul and score. Only modernjev-decide answers rank and route. modernjev-decide is the smaller model, at 150M parameters to 395M. bekko-system-one is licensed other; modernjev-decide, apache-2.0.
Only modernjev-decide publishes an accuracy figure (73.4% on AgentToolDecisions-180K held-out test, next action type (1,158 decisions, 3 choices; the maker's split)); bekko-system-one does not, so there is no comparison to make without your own test.
bekko-system-one: Free (open weights). modernjev-decide: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull hotchpotch/bekko-system-one and systemone pull maziyar-panahi/modernjev-decide download the weights.