# Contrastive-LM: clm

> 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.

- Page: https://systemonemodels.ai/contrastive-lm/clm
- API: https://api.systemonemodels.ai/v1/models/contrastive-lm/clm
- Download: `pip install systemonemodels && systemone pull contrastive-lm/clm`

## Facts

| | |
|---|---|
| Maker | Contrastive-LM (https://systemonemodels.ai/contrastive-lm) |
| Decides | choice, score, noul, rank, classify, route |
| Architecture | clm |
| Base model | qwen/qwen3-8b |
| Parameters | 8.0B |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-23 |
| Latest version | 2026.09.25 |

## Model card

<!-- generated by scripts/seed_catalog.py; edit content/models/catalog.yaml -->

# CLM v0.1-8B

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.

The repository holds only the two heads (about 75 MB); the frozen Qwen3-8B encoder is fetched separately and needs a vLLM pooling server on Linux with an NVIDIA GPU. Training used roughly 60M question–answer pairs, 30M hard negatives and 1M agent trajectories. The authors claim parity with Jev on computer-use, gaming and tool-calling, up to 9× lower latency, and, after fine-tuning the heads, 87.6% on Terminal-Bench 2.1 and 81.6% on DeepSWE; none of these are independently reproduced. The server exposes /v1/systemone and /v1/rank; confidence is the top probability minus the mean of the rest, and states are rendered as prose. A multimodal CLM-35B is announced for October.

## What it decides

- **choice** — picks one option from a set
- **score** — places the input on an ordered scale
- **noul** — answers a yes/no question with one calibrated probability
- **rank** — orders a set of items
- **classify** — assigns a category from a fixed taxonomy
- **route** — sends the input to one of several destinations

## At a glance

| | |
|---|---|
| Parameters | 8B |
| Base model | `qwen/qwen3-8b` |
| Maker | Contrastive-LM |
| Released | 2026-09-23 |
| License | apache-2.0 |
| Reported latency | authors: up to 9× lower than Jev; ~28 ms for a new state on an RTX 4090 |

## Get the weights

```bash
pip install systemonemodels
systemone pull contrastive-lm/clm
```

The files are served from the maker's Hugging Face repository, [`Contrastive-LM/CLM-v0.1-8B`](https://huggingface.co/Contrastive-LM/CLM-v0.1-8B), and verified against the checksums recorded here.

## Read more

- [Source and API reference](https://github.com/Contrastive-LM/CLM)
- [Weights on Hugging Face](https://huggingface.co/Contrastive-LM/CLM-v0.1-8B)

---

*This page was opened by System One for Contrastive-LM, who can claim the organisation and take it over at any time.*

---

From System One Models — https://systemonemodels.ai/ · every System One model: https://systemonemodels.ai/system-one-models