Loading comparison…
HEAD TO HEAD
Nokia Applied Research: anyjev and Contrastive-LM: clm, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | nokia-applied-research/anyjev | contrastive-lm/clm |
|---|---|---|
| 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. | Contrastive Language Models score a state against a set of candidate actions with a contrastive objective. Two projection heads on frozen Qwen3-8B embeddings, trained with InfoNCE; clm-serve maps Choice, Score and Noul onto candidate ranking. |
| Decides | choice, score, noul, classify, route | choice, score, noul, rank, classify, route |
| Architecture | anyjev | clm |
| Fine-tuned from | qwen/qwen3-8b | qwen/qwen3-8b |
| License | apache-2.0 | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | 77.1% | — |
| Calibration error | 0.034 | — |
| Valid action rate | — | — |
| Median latency | — | — |
| p95 latency | — |
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 Applied Research and clm from Contrastive-LM. Both have open weights you can download and run. Both answer choice, score, noul, classify and route questions. Only clm answers rank.
Only anyjev publishes an accuracy figure (77.1% on LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question); clm does not, so there is no comparison to make without your own test.
anyjev: Free (open weights). clm: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull nokia-applied-research/anyjev and systemone pull contrastive-lm/clm download the weights.
| — |
| Evaluation suite | LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question | — |
|---|
| Latest version | 0.1.0 | 2026.09.25 |
|---|
| Variants | — | LICENSE |
|---|
| Size of latest version | — | 72.1 MB |
|---|
| Files | 0 | 3 |
|---|
| Downloads | 0 | 0 |
|---|
| Stars | 0 | 0 |
|---|
| Tags | system-one, qwen, training-free, calibrated, vllm, 8b | system-one, contrastive, qwen, ranking, 8b |
|---|
| Updated | Sep 27, 2026 | Sep 27, 2026 |
|---|