HEAD TO HEAD
Contrastive-LM: clm and Kotoba Labs: open-jev-deberta-v3-large, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | contrastive-lm/clm | kotoba-labs/open-jev-deberta-v3-large |
|---|---|---|
| Summary | 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. | An independent Jev-shaped reproduction on DeBERTa-v3-large. One state and any number of choice, score and yes/no questions go in, and a calibrated distribution per question comes out of one forward pass. Public gold labels only. |
| Decides | choice, score, noul, rank, classify, route | choice, score, noul, classify |
| Architecture | clm | open-jev |
| Fine-tuned from | qwen/qwen3-8b | microsoft/deberta-v3-large |
| License | apache-2.0 | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | — | 85.4% |
| Calibration error | — | 0.022 |
| Valid action rate | — | — |
| Median latency | — | 28 ms |
| p95 latency | — | — |
| Evaluation suite | — | Kotoba held-out test, seen question types (banking77, SST-5, BoolQ) |
| Latest version | 2026.09.25 | 2026.09 |
| Variants | LICENSE | — |
| Size of latest version | 72.1 MB | 1.6 GB |
| Files | 3 | 9 |
| Downloads | 43 | 2 |
| Stars | 0 | 0 |
| Tags | system-one, contrastive, qwen, ranking, 8b | system-one, encoder, deberta, public-data, 434m |
| Updated | Nov 19, 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.
clm is from Contrastive-LM and open-jev-deberta-v3-large from Kotoba Labs. Both have open weights you can download and run. Both answer choice, score, noul and classify questions. Only clm answers rank and route. open-jev-deberta-v3-large is the smaller model, at 434M parameters to 8.0B.
Only open-jev-deberta-v3-large publishes an accuracy figure (85.4% on Kotoba held-out test, seen question types (banking77, SST-5, BoolQ)); clm does not, so there is no comparison to make without your own test.
clm: Free (open weights). open-jev-deberta-v3-large: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull contrastive-lm/clm and systemone pull kotoba-labs/open-jev-deberta-v3-large download the weights.