Clef-omni is Cloudflare's decision model for audio and video on Workers AI. It has 30B parameters with 3B active per request.
By Cloudflare · Released 9 Oct 2026
The Clef family on Workers AI answers typed questions in one pass. Clef-omni adds audio and video, so it can decide about what is seen and heard at the same time. The same update cut Clef-flash to $0.038 per million input tokens.
Clef-omni: Published price (vendor-reported): $0.150 per million input tokens
Clef-omni: Cloudflare
Source: Cloudflare changelog →
A live site where OpenAI Decisions, Jev and Cloudflare's Clef put a probability on live questions after reading recent news, never the odds. Their probabilities are compared against Polymarket and Kalshi prices.
A Chrome extension: describe what you don't want to see in plain words, and eligible posts on your X timeline get a probability per rule. By default, anything over your threshold folds into a one-line bar you can open again.
Clef scores tiles from Jamen Percy's 2026 Burning Man gigapan, then the code sums expected RV counts, car counts and shade coverage. The author's 256 px run judged 8,785 tiles for $5.29, estimating 14,385 RVs, 18,084 cars and 256.8 acres of shade. These are estimates: vehicles under shade are hidden, and dense 512 px tiles undercount. Splitting validation tiles into 256 px quadrants brought RV estimates to 386 against 394 from the LLM counters.
Author's recorded runs · VENDOR-REPORTED
Three-label sentiment code and saved predictions for Jev, Clef and Clef-flash on eight European languages. The zero-shot report covers 12,970 test items. Its comparison with five LLMs uses published paper scores with a different prompt format, rather than fresh matched calls. Broader prepared datasets and tuning runs are separate from this reported scope.
Author-run records; no fresh model inference reproduced
Source: Licence → Source: Zero-shot report → Source: Saved score table → Source: Jev English predictions → Source: Inference code →
Moves llama.cpp decision serving code into a C library. Its recorded d1 request takes 678 ms on eight CPU threads. Saved CLI/server parity tests stay within 0.01 for Laya, exceed that probability tolerance for H2O and Clef, and reject the tested Perplexity package. On a Pixel 8a in Termux, the maker reports a 464 ms yes/no decision with d1-2B Q4_K_M on four threads and 2.1 GB RSS. Without repacking, that question takes 1,210 ms at 1.1 GB RSS. iOS and Capacitor app bindings remain plans.
Source: Licence → Source: Saved parity output → Source: Phone measurement and build notes →