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.
A rank-16 LoRA on Qwen3.5-4B built for the JevBench setting. A document, a policy and a question go in; a calibrated distribution over the option letters comes out of one forward pass. Research and demo use only.
Decides
choice, score, noul, rank, classify, route
choice, score, noul, classify
Architecture
clm
hopper
Fine-tuned from
qwen/qwen3-8b
qwen/qwen3.5-4b
License
apache-2.0
Research and demo use only (training data includes RACE, non-commercial); serving code Apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
68.5%
Calibration error
—
0.102
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.
Questions
What is the difference between clm and hopper?
clm is from Contrastive-LM and hopper from HopitAI. Both have open weights you can download and run. Both answer choice, score, noul and classify questions. Only clm answers rank and route. hopper is the smaller model, at 4.0B parameters to 8.0B. clm is licensed apache-2.0; hopper, other.
Which is more accurate, clm or hopper?
Only hopper publishes an accuracy figure (68.5% on JevBench public hard tier (111 items, measured by the authors)); clm does not, so there is no comparison to make without your own test.
Which is cheaper, clm or hopper?
clm: Free (open weights). hopper: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run clm or hopper locally?
Yes, both: systemone pull contrastive-lm/clm and systemone pull hopit-ai/hopper download the weights.
—
Evaluation suite
—
JevBench public hard tier (111 items, measured by the authors)
