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.
Qwen3.5-4B with a merged rank-16 LoRA distilled from 17,408 teacher questions and 30,000 public training rows, read out as a temperature-scaled softmax over answer-letter logits. Also in 9B, 2B and GGUF.
Decides
choice, score, noul, rank, classify, route
choice, score, noul, classify, route
Architecture
clm
jevk5
Fine-tuned from
qwen/qwen3-8b
qwen/qwen3.5-4b
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
78.4%
Calibration error
—
0.035
Valid action rate
—
—
Median latency
—
13.2 ms
p95 latency
—
—
Evaluation suite
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.
Questions
What is the difference between clm and jevk5?
clm is from Contrastive-LM and jevk5 from Alibi Serikbay. Both have open weights you can download and run. Both answer choice, score, noul, classify and route questions. Only clm answers rank. jevk5 is the smaller model, at 4.0B parameters to 8.0B.
Which is more accurate, clm or jevk5?
Only jevk5 publishes an accuracy figure (78.4% on JevBench public hard tier); clm does not, so there is no comparison to make without your own test.
Which is cheaper, clm or jevk5?
clm: Free (open weights). jevk5: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run clm or jevk5 locally?
Yes, both: systemone pull contrastive-lm/clm and systemone pull alibi-serikbay/jevk5 download the weights.
