# Clef-omni

Decision models · checked 2026-10-14 · https://dbggr.dev/models/clef-omni/

Clef-omni is an open-weight decision model from Cloudflare (Apache 2.0, 30B MoE (3B active), on Qwen3-Omni-30B-A3B parameters): it answers typed questions, such as yes or no, pick one or pick several, with probabilities instead of writing text. It costs $0.15 per million input tokens (Cloudflare Workers AI, OpenRouter), and output is free. It scores 58.3 (self-hosted; seconds are the board's estimate; not capability-ranked, as its cost is above the board's cap) on the independent JevBench Capability Score (board v1.7.46), where Jev, the reference, scores 77.1.

- Made by: Cloudflare
- Weights: Open, Apache 2.0
- Size: 30B MoE (3B active), on Qwen3-Omni-30B-A3B
- Where: Cloudflare Workers AI, OpenRouter
- Images: Yes
- Context: 64,000
- Fine-tuning: Not stated
- Docs: https://developers.cloudflare.com/workers-ai/models/clef-omni/

## Clef-omni 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 |
|---|---|---|---|---|
| Cloudflare Workers AI, OpenRouter | $0.15 | Free | Output is free. Also takes audio (about 780 tokens a minute) and video (about 8,600 tokens a minute at 480p, plus the audio track), billed as input tokens. | [2026-10-14](https://blog.cloudflare.com/clef-faster-cheaper-multimodal/) |

## Clef-omni benchmarks

| Benchmark | Result | What it measures | Reported by |
|---|---|---|---|
| JevBench v1.7.46 | 58.3 | Capability Score: accuracy and calibration on typed decisions, with Jev as the reference system | [Independent](https://benchmarkheaven.com/jev-models) |
| JevBench time | 0.44 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 | About 130 ms median for text, 150 ms for images, 1.5 s for a 21-second video with sound | Latency on the maker's own hardware | Cloudflare |

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

## Decision models like Clef-omni

| Model | Input / M | JevBench | Weights |
|---|---|---|---|
| Clef-omni | $0.15 | 58.3 | 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 |
| [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 |
| [CLM-8B](https://dbggr.dev/models/clm-8b/) | Self-host | 20.1 | Open |
| [Strands Decider 2B](https://dbggr.dev/models/strands-decider-2b/) | Self-host | – | Open |

## FAQ

### How much does Clef-omni cost?

Clef-omni costs $0.15 per million input tokens at Cloudflare Workers AI, OpenRouter. Output is free. Also takes audio (about 780 tokens a minute) and video (about 8,600 tokens a minute at 480p, plus the audio track), billed as input tokens. Checked 14 October 2026.

### Is Clef-omni open weight?

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

### How fast is Clef-omni?

Cloudflare reports About 130 ms median for text, 150 ms for images, 1.5 s for a 21-second video with sound, measured on its own hardware. JevBench gives 0.44 seconds per decision, its estimate for self-hosting.

### How does Clef-omni compare with Jev?

On JevBench's Capability Score (board v1.7.46), Clef-omni scores 58.3 against Jev's 77.1. Jev costs $0.042 per million input tokens; Clef-omni costs $0.15. JevBench is a one-person project that uses Jev as its reference, which favours cheap hosted models.