# CLM-8B

Decision models · checked 2026-10-14 · https://dbggr.dev/models/clm-8b/

CLM-8B is an open-weight decision model from Stanford and NVIDIA (Apache 2.0, 8B parameters): it answers typed questions, such as yes or no, pick one or pick several, with probabilities instead of writing text. There's no hosted version: you run it on your own hardware. It scores 20.1 (self-hosted; seconds are the board's estimate) on the independent JevBench Capability Score (board v1.7.46), where Jev, the reference, scores 77.1.

- Made by: Stanford and NVIDIA
- Weights: Open, Apache 2.0
- Size: 8B
- Where: Self-host only
- Images: No, text only
- Context: 2,048
- Fine-tuning: Tools provided
- Docs: https://huggingface.co/Contrastive-LM/CLM-v0.1-8B

## CLM-8B price

US dollars per million input tokens. Decision models don't write text, so none of the hosted ones charge for output.

| Where | Input | Output | Notes | Checked |
|---|---|---|---|---|
| Self-host only | Self-host | – | – | [2026-10-14](https://huggingface.co/Contrastive-LM/CLM-v0.1-8B) |

## CLM-8B benchmarks

| Benchmark | Result | What it measures | Reported by |
|---|---|---|---|
| JevBench v1.7.46 | 20.1 | Capability Score: accuracy and calibration on typed decisions, with Jev as the reference system | [Independent](https://benchmarkheaven.com/jev-models) |
| JevBench time | 0.19 s | Seconds per decision: measured end to end for hosted APIs, the board's estimate for self-hosted models | [Independent](https://benchmarkheaven.com/jev-models) |
| Reported speed | Up to 9× faster than Jev | Latency on the maker's own hardware | Stanford and NVIDIA |

Makers' speeds are measured on their own hardware and aren't comparable with each other or with JevBench's timings.

## Decision models like CLM-8B

| Model | Input / M | JevBench | Weights |
|---|---|---|---|
| CLM-8B | Self-host | 20.1 | Open |
| [Jev 1.13](https://dbggr.dev/models/jev/) | $0.042 | 77.1 | Closed |
| [Clef](https://dbggr.dev/models/clef/) | $0.24 | 75.3 | Open |
| [d1](https://dbggr.dev/models/d1/) | $0.04 | 74.4 | Closed |
| [OpenAI Decisions (GPT-6 Luna)](https://dbggr.dev/models/openai-decisions/) | $0.10 | 73.5 | Closed |
| [Microsoft-Decision-1](https://dbggr.dev/models/microsoft-decision-1/) | $0.042 | 70.8 | Closed |
| [Clef-flash](https://dbggr.dev/models/clef-flash/) | $0.038 | 63.8 | Open |
| [Clef-omni](https://dbggr.dev/models/clef-omni/) | $0.15 | 58.3 | Open |
| [d1-3B](https://dbggr.dev/models/d1-3b/) | Self-host | 49.5 | Open |
| [d1-omni-600M (experimental)](https://dbggr.dev/models/d1-omni-600m/) | Self-host | 32.8 | Open |
| [Laya](https://dbggr.dev/models/laya/) | Free | 32.5 | Open |
| [Strands Decider 2B](https://dbggr.dev/models/strands-decider-2b/) | Self-host | – | Open |

## FAQ

### How much does CLM-8B cost?

CLM-8B has no hosted, per-token price: you download it and run it yourself, so the cost is your own compute.

### Is CLM-8B open weight?

Yes. Stanford and NVIDIA publishes the weights under Apache 2.0, so you can run and fine-tune it yourself.

### How fast is CLM-8B?

Stanford and NVIDIA reports Up to 9× faster than Jev, measured on its own hardware. JevBench gives 0.19 seconds per decision, its estimate for self-hosting.

### How does CLM-8B compare with Jev?

On JevBench's Capability Score (board v1.7.46), CLM-8B scores 20.1 against Jev's 77.1. Jev costs $0.042 per million input tokens; CLM-8B is self-host only. JevBench is a one-person project that uses Jev as its reference, which favours cheap hosted models.