Cloudflare Open-Sources Decision Models Clef, Now on Ollama
Cloudflare open-sourced its decision models Clef (27B) and Clef Flash (9B), fine-tuned from Qwen3.5 and now available on Ollama. The multimodal models support image classification and ticket routing, with a user reporting just 53ms per decision on an RTX 5090.
2026-10-03 ~ 2026-10-04 · 3 related posts
- Episode 1: Cloudflare Open-Sources Its First Models clef and clef-flash, Topping the Jev Decision Index(2026-10-01, 23 posts)
- Episode 2: Cloudflare Launches RL Fine-Tuning with Open Decision Model(2026-10-01, 2 posts)
- Episode 3: Cloudflare Open-Sources Decision Models Clef, Now on Ollama(2026-10-03, 3 posts)
- Episode 4: Cloudflare Launches Omni-Modal Clef-omni and Cuts Clef-flash Price(2026-10-10, 2 posts)
- Episode 5: Cloudflare open-sources Clef decision models, topping the Jev decision index(2026-10-10, 2 posts)
- Episode 6: Clef overtakes OpenAI to rank second in decision model market share(2026-10-12, 2 posts)
- Cloudflare's Clef decision models (27B and 9B) now available on Ollama — ollama · 2026-10-03
- Clef Flash: 9B multimodal decision model fine-tuned from Qwen3.5, 256K context — ollama · 2026-10-03
- Cloudflare's open-weights decision model Clef scores all answers in one pass, 53ms per decision on RTX 5090 — michellechen · 2026-10-04