Loading comparison…
HEAD TO HEAD
Yuichi Tateno (hotchpotch): bekko-system-one and Superagent: security-one, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | hotchpotch/bekko-system-one | superagent/security-one |
|---|---|---|
| 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. | Superagent's 27B security decision model. It scores 2 to 16 supplied options in one forward pass and returns a probability for each, for triaging prompt injection, agent tool calls, code changes, logs and alerts. Apache-2.0 weights (gated, auto-approved) plus a hosted API. |
| Decides | choice, noul, score | choice, score, noul, classify, route |
| Architecture | bekko | security-one |
| Fine-tuned from | cross-encoder/ettin-reranker-400m-v1 | denis-pplx/autojev-27b |
| License | Unassigned: the weights licence is not finalised (treated as non-commercial); training code MIT | apache-2.0 |
| Availability | Open weights | Open weights + hosted API |
| Hosted by | — | Superagent |
| Input price | — | — |
| Decision accuracy | — | 99.8% |
| 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) | BIPIA binary prompt-injection detector adaptation (800 rows; Superagent's frozen evaluation, unsafe if P(unsafe) >= 0.70) |
| Latest version | 2026.09 | 2026.09 |
| Variants | 0_BekkoInference | — |
| Size of latest version | 1.5 GB | — |
| Files | 9 | 0 |
| Downloads | 0 | 0 |
| Stars | 0 | 0 |
| Tags | system-one, encoder, ettin, onnx, browser, small, 395m | system-one, security, prompt-injection, qwen, sglang, hosted, gated, 27b |
| 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 security-one from Superagent. bekko-system-one has open weights you can download and run; security-one has open weights and a hosted API. Both answer choice, noul and score questions. Only security-one answers classify and route. bekko-system-one is the smaller model, at 395M parameters to 27B. bekko-system-one is licensed other; security-one, apache-2.0.
Only security-one publishes an accuracy figure (99.8% on BIPIA binary prompt-injection detector adaptation (800 rows; Superagent's frozen evaluation, unsafe if P(unsafe) >= 0.70)); bekko-system-one does not, so there is no comparison to make without your own test.
bekko-system-one: Free (open weights). security-one: Hosted, price not published, or free to self-host. Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull hotchpotch/bekko-system-one and systemone pull superagent/security-one download the weights.