All models

COMPARE

Put models side by side.

What each one decides, how well calibrated it is, how fast it answers and what it costs to pull. Up to 4 at a time; the better value in each row is marked.

Propertycontrastive-lm/clm
SummaryContrastive 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.
Decideschoice, score, noul, rank, classify, route
Architectureclm
Fine-tuned fromqwen/qwen3-8b
Licenseapache-2.0
AvailabilityOpen weights
Hosted by—
Input price—
Decision accuracy—
Calibration error—
Valid action rate—
Median latency—
p95 latency—
Evaluation suite—
Latest version2026.09.25
VariantsLICENSE
Size of latest version72.1 MB
Files3
Downloads0
Stars0
Tagssystem-one, contrastive, qwen, ranking, 8b
UpdatedSep 25, 2026