HEAD TO HEAD
Yuichi Tateno (hotchpotch): bekko-system-one and Dimitris Papadopoulos: biodecision, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | hotchpotch/bekko-system-one | lighteternal/biodecision |
|---|---|---|
| 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 4B biomedical decision model by Dimitris Papadopoulos: Qwen3.5-4B tuned with Together AI's Tev1 recipe on 1.75M pharma, clinical-trial and literature decisions. One forward pass gives a temperature-scaled probability per option. Research use only. |
| Decides | choice, noul, score | choice, score, noul, classify |
| Architecture | bekko | biodecision |
| Fine-tuned from | cross-encoder/ettin-reranker-400m-v1 | qwen/qwen3.5-4b |
| License | Unassigned: the weights licence is not finalised (treated as non-commercial); training code MIT | Research use only (several training sources are non-commercial); not validated for patient care |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | — | 75.3% |
| Calibration error | — | — |
| Valid action rate | — | — |
| Median latency | 5 ms | 39 ms |
| p95 latency | 5.4 ms | — |
| Evaluation suite | The maker's short-input speed probe (4-option Choice, RTX 5090, 1 Oct 2026) | BioDecision v2 held-out test set (63,432 overlap-screened decisions from public biomedical benchmarks; the maker's own split) |
| Latest version | 2026.09 | 2.0.0 |
| Variants | 0_BekkoInference | LICENSE |
| Size of latest version | 1.5 GB | 8.7 GB |
| Files | 9 | 15 |
| Downloads | 0 | 0 |
| Stars | 0 | 0 |
| Tags | system-one, encoder, ettin, onnx, browser, small, 395m | system-one, qwen, biomedical, clinical-trials, jev-inspired, research-only, 4b |
| 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 biodecision from Dimitris Papadopoulos. Both have open weights you can download and run. Both answer choice, noul and score questions. Only biodecision answers classify. bekko-system-one is the smaller model, at 395M parameters to 4.0B.
Only biodecision publishes an accuracy figure (75.3% on BioDecision v2 held-out test set (63,432 overlap-screened decisions from public biomedical benchmarks; the maker's own split)); bekko-system-one does not, so there is no comparison to make without your own test.
bekko-system-one: Free (open weights). biodecision: 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 biodecision in about 39 ms — measured on different hardware, so treat it as a rough guide.
Yes, both: systemone pull hotchpotch/bekko-system-one and systemone pull lighteternal/biodecision download the weights.