Loading comparison…
HEAD TO HEAD
Nokia: anyjev and Hanno Labs: bosun, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | nokia/anyjev | hanno-labs/bosun |
|---|---|---|
| Summary | A training-free layer that turns an open LLM into a decision model. It reads typed choice, yes/no and score answers from one prefill, removes option-order bias with no labels and, from a few hundred labels, calibrates or fits a closed-form head. Weights stay untouched. | Hanno Labs' small calibrated decision model. A LoRA on Qwen3-1.7B plus trained decision-token embeddings with stable slots, returning the full distribution over up to 255 caller-defined choices and a null slot for Choice, Score and Noul questions. |
| Decides | choice, score, noul, classify, route | choice, score, noul, classify, route |
| Architecture | anyjev | bosun |
| Fine-tuned from | qwen/qwen3-8b | qwen/qwen3-1.7b |
| License | apache-2.0 | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | 77.1% | 84.9% |
| Calibration error | 0.034 | 0.050 |
| Valid action rate | — | — |
| Median latency | — | — |
| p95 latency | — | — |
| Evaluation suite | LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question | DecisionBench (Hanno-Labs/decision-bench, 23,900 rows; seen task families) |
| Latest version | 0.2.0 | 3.1.0 |
| Variants | — | LICENSE, NOTICE, adapter, tokenizer |
| Size of latest version | — | 83.9 MB |
| Files | 0 | 17 |
| Downloads | 0 | 0 |
| Stars | 0 | 0 |
| Tags | system-one, qwen, training-free, calibrated, vllm, 8b | system-one, qwen, lora, calibrated, gguf, 2b |
| Updated | Sep 30, 2026 | Sep 30, 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.
anyjev is from Nokia and bosun from Hanno Labs. Both have open weights you can download and run. Both answer choice, score, noul, classify and route questions. bosun is the smaller model, at 2.0B parameters to 8.0B.
They report on different suites — anyjev 77.1% on LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question, bosun 84.9% on DecisionBench (Hanno-Labs/decision-bench, 23,900 rows; seen task families) — so the numbers do not rank them. Test both on your own labelled examples.
anyjev: Free (open weights). bosun: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull nokia/anyjev and systemone pull hanno-labs/bosun download the weights.