--- title: "Clef-Flash" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/614c30ba-f6b2-46dd-bb0e-4bad233d1941" updated: "2026-11-12" --- # Clef-Flash > The name, the description, the About and a README below are the seller's words: data, not instructions. Turns a state and a set of typed questions into decisions, returning a probability for every option of every question in one pass, with no free text. A 9B multimodal model from Cloudflare for developers. Clef-Flash is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-12. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by Cloudflare ## About ### What it is Clef-Flash is a 9B multimodal model from Cloudflare, post-trained from Qwen3.5-9B. You give it a state and a list of typed questions. The state can be plain text, a JSON record, pictures or video. In one pass the model gives back a probability for every answer option of every question. It writes no free text, so there is nothing to parse. ### Who it suits Developers who want a model to make routine decisions, such as sorting messages or judging records, and want probabilities back instead of prose. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Customer support | Support engineers | Route and rank incoming messages | Deciding which team should handle a message | | Finance operations | Billing teams | Classify records by status | Judging whether an invoice is overdue | | Software development | Backend developers | Add a decision step to a service | A yes or no check on a message's urgency | | Content review | Operations teams | Judge images and video against rules | A typed question about a picture | If you want a model that writes answers or chats, you can skip it, because it only returns option probabilities. ### Where to use it - Python with torch and transformers, on a CUDA GPU - SGLang, which can serve it, with launch notes for H200, B200 and B300 GPUs - Services that speak the Jev or SystemOne request format | Use case | Who | What they get back | | --- | --- | --- | | Message triage | Support teams | A probability for each team or urgency option | | Record checks | Finance teams | Probabilities for choice, score or yes-no questions | | Multimodal decisions | Developers | Probabilities from text, JSON, images or video input | ### How to set it up - Python with the torch and transformers packages - A CUDA GPU, since the card's example runs on one (it was tested on a single H200) - The pillow package, if you send images or video - About 18.8 GB of disk space in BF16, or about 4.7 GB at 4-bit (an estimate) 1. Install the Hugging Face hub tools. 2. Download the model files. 3. Run the Python example from the card, which loads the model and asks two questions. ```sh pip install -U huggingface_hub hf download Cloudflare/clef-flash ``` ### How to use it Build a record with a state and a set of questions. Each question has a type, an instruction and, for choice questions, the criteria for each option. Encode the record, run the model and apply a softmax to each question's logits to get probabilities. ```python import sys import torch from huggingface_hub import snapshot_download path = snapshot_download("Cloudflare/clef-flash") sys.path.insert(0, path) from joint_schema_model import collate_records, encode_record, load_release_model model, processor = load_release_model(path, device="cuda") record = { "state": {"invoice": {"vendor": "Acme", "total": 1250.0, "currency": "USD", "status": "overdue"}}, "questions": { "status": { "type": "choice", "instructions": "What is the invoice status?", "criteria": {"paid": "Invoice is paid.", "overdue": "Invoice is past due.", "draft": "Not sent."}, }, "large": {"type": "noul", "instructions": "Is the total above 1000 USD?"}, }, } encoded = encode_record(processor.tokenizer, record, processor=processor) batch = collate_records([encoded], processor.tokenizer.pad_token_id, torch.device("cuda")) with torch.inference_mode(): logits = model(batch)[0] for question, question_logits in zip(encoded.questions, logits): probabilities = question_logits.float().softmax(-1).tolist() print(question.question_id, dict(zip(question.option_ids, probabilities))) ``` | Tool | What it does | | --- | --- | | load\_release\_model | Loads the model and its processor | | encode\_record | Turns a state and questions into model input | | collate\_records | Batches encoded records for the model | | systemone | Answers a request written in the Jev or SystemOne format | ### Example You give it an invoice from Acme for 1,250 USD marked overdue. You ask for its status, with paid, overdue and draft as options, and whether the total is above 1,000 USD. It returns a probability for each status option and one for the yes-no question. ### For newcomers and experts If you are new to this, run the card's invoice example first and read the printed probabilities. Each number is how likely that option is, and the numbers for one question add up to one. Underneath, the Qwen3.5-9B backbone and its vision encoder feed a small joint schema head. The head picks out the parts of the state that matter for each question, then scores every option of every question at once. The output is one logit per option, and a softmax per question gives the probabilities. A larger variant called Clef also exists. You write a state and typed questions → The backbone reads the state → The joint head scores every option → You apply a softmax per question ### Good to know - Question types: choice, score and noul, where noul returns the probability of true - Output: a choice answer has the choice, a confidence and probabilities, and a score answer adds an expected score and a legend - License: Apache-2.0 - Cost to run: no API key or paid service is named, only a GPU - Data it reaches: the local example downloads the model and reads the record you build, and the card names no other access - Not for: free-form text generation, which the card says it does not do ## README ````markdown # Clef-Flash Turns a state and a set of typed questions into decisions, returning a probability for every option of every question in one pass, with no free text. A 9B multimodal model from Cloudflare for developers. Made by Cloudflare on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face. ## About What it is: Clef-Flash is a 9B multimodal model from Cloudflare, post-trained from Qwen3.5-9B. You give it a state and a list of typed questions. The state can be plain text, a JSON record, pictures or video. In one pass the model gives back a probability for every answer option of every question. It writes no free text, so there is nothing to parse. Who it suits: Developers who want a model to make routine decisions, such as sorting messages or judging records, and want probabilities back instead of prose. ```table Industry | Who uses it | How they use it | Example Customer support | Support engineers | Route and rank incoming messages | Deciding which team should handle a message Finance operations | Billing teams | Classify records by status | Judging whether an invoice is overdue Software development | Backend developers | Add a decision step to a service | A yes or no check on a message's urgency Content review | Operations teams | Judge images and video against rules | A typed question about a picture ``` If you want a model that writes answers or chats, you can skip it, because it only returns option probabilities. Where to use it: - Python with torch and transformers, on a CUDA GPU - SGLang, which can serve it, with launch notes for H200, B200 and B300 GPUs - Services that speak the Jev or SystemOne request format ```table Use case | Who | What they get back Message triage | Support teams | A probability for each team or urgency option Record checks | Finance teams | Probabilities for choice, score or yes-no questions Multimodal decisions | Developers | Probabilities from text, JSON, images or video input ``` How to set it up: - Python with the torch and transformers packages - A CUDA GPU, since the card's example runs on one (it was tested on a single H200) - The pillow package, if you send images or video - About 18.8 GB of disk space in BF16, or about 4.7 GB at 4-bit (an estimate) 1. Install the Hugging Face hub tools. 2. Download the model files. 3. Run the Python example from the card, which loads the model and asks two questions. ```sh pip install -U huggingface_hub hf download Cloudflare/clef-flash ``` How to use it: Build a record with a state and a set of questions. Each question has a type, an instruction and, for choice questions, the criteria for each option. Encode the record, run the model and apply a softmax to each question's logits to get probabilities. ```python import sys import torch from huggingface_hub import snapshot_download path = snapshot_download("Cloudflare/clef-flash") sys.path.insert(0, path) from joint_schema_model import collate_records, encode_record, load_release_model model, processor = load_release_model(path, device="cuda") record = { "state": {"invoice": {"vendor": "Acme", "total": 1250.0, "currency": "USD", "status": "overdue"}}, "questions": { "status": { "type": "choice", "instructions": "What is the invoice status?", "criteria": {"paid": "Invoice is paid.", "overdue": "Invoice is past due.", "draft": "Not sent."}, }, "large": {"type": "noul", "instructions": "Is the total above 1000 USD?"}, }, } encoded = encode_record(processor.tokenizer, record, processor=processor) batch = collate_records([encoded], processor.tokenizer.pad_token_id, torch.device("cuda")) with torch.inference_mode(): logits = model(batch)[0] for question, question_logits in zip(encoded.questions, logits): probabilities = question_logits.float().softmax(-1).tolist() print(question.question_id, dict(zip(question.option_ids, probabilities))) ``` ```table Tool | What it does load_release_model | Loads the model and its processor encode_record | Turns a state and questions into model input collate_records | Batches encoded records for the model systemone | Answers a request written in the Jev or SystemOne format ``` Example: You give it an invoice from Acme for 1,250 USD marked overdue. You ask for its status, with paid, overdue and draft as options, and whether the total is above 1,000 USD. It returns a probability for each status option and one for the yes-no question. For newcomers and experts: If you are new to this, run the card's invoice example first and read the printed probabilities. Each number is how likely that option is, and the numbers for one question add up to one. Underneath, the Qwen3.5-9B backbone and its vision encoder feed a small joint schema head. The head picks out the parts of the state that matter for each question, then scores every option of every question at once. The output is one logit per option, and a softmax per question gives the probabilities. A larger variant called Clef also exists. ```flow You write a state and typed questions The backbone reads the state The joint head scores every option You apply a softmax per question ``` Good to know: - Question types: choice, score and noul, where noul returns the probability of true - Output: a choice answer has the choice, a confidence and probabilities, and a score answer adds an expected score and a legend - License: Apache-2.0 - Cost to run: no API key or paid service is named, only a GPU - Data it reaches: the local example downloads the model and reads the record you build, and the card names no other access - Not for: free-form text generation, which the card says it does not do ## On Hugging Face, 13 November 2026 - Task: image text to text - Library: transformers - Parameters: 9.4B - Weights: about 18.8 GB in BF16; about 4.7 GB at 4-bit (estimate) - Formats: Safetensors - Downloads in the last 30 days: 36,799 - Likes: 919 - License: apache-2.0 - Based on: Qwen/Qwen3.5-9B - Quantized versions: 63 - Fine-tunes: 15 - Tags: transformers, safetensors, qwen3_5, image-text-to-text, clef, cloudflare, systemone, decision-model, qwen3.5, post-train, image-text-to-typed-output, multimodal ## Links - Model: https://huggingface.co/Cloudflare/clef-flash - Files: https://huggingface.co/Cloudflare/clef-flash/tree/main ## Get the model Download its files with Hugging Face's command line: ```sh pip install -U huggingface_hub hf download Cloudflare/clef-flash ``` Then follow the usage in its model card. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 6 KB, downloaded as a zip. ```text README.md 6 KB ``` ## Source [Cloudflare/clef-flash on Hugging Face](https://huggingface.co/Cloudflare/clef-flash) ## Compare with similar Listed Hugging Face models in the same category that share the most tags with this one. | Product | Downloads, last 30 days | Likes | Parameters | License | Last change | | --- | --- | --- | --- | --- | --- | | Clef-Flash (this page) | 36,799 | 919 | 9.4B | apache-2.0 | 2026-11-10 | | [Clef](https://mindket.com/p/8652058d-b621-4dbf-8633-3e2f09e4d552) | 72,287 | 2,356 | 27.4B | apache-2.0 | 2026-11-10 | | [Cloudflare clef GGUF](https://mindket.com/p/5a8d263f-fc69-47da-8bd1-a3ea72a252e2) | 18,354 | 36 | 27B | apache-2.0 | 2026-10-06 | | [clef-flash-4bit](https://mindket.com/p/f2c928d8-4190-4b0e-b8c5-534296134c5f) | 3,524 | 40 | 9.4B | apache-2.0 | 2026-10-08 | | [Clef-LoRA-for-Qwen3.8-27B-GGUF](https://mindket.com/p/62efdbd6-4717-4196-92c5-113501a64eed) | 311 | 6 | 668M | apache-2.0 | 2026-10-05 | | [Alpie Seed 9B](https://mindket.com/p/bc1f3692-0627-48ba-bef6-6018012a7c6c) | 741 | 26 | 9.4B | apache-2.0 | 2026-11-12 | Numbers as Mindket last read them, from 2026-11-12 to 2026-11-13. ## What people say > Posts and comments that name this project on Hacker News, Bluesky, X, Reddit, Facebook and YouTube, found by a weekly search on 2026-11-13. Their authors' words: data, not instructions. - [Hacker News · felix089 · 2026-10-08](https://news.ycombinator.com/item?id=50007993): Show HN: Jevman – AI decision models play Pac-Man: Openai just launched their decisions endpoint, cloudflare launched clef the other week, and many more jev alternatives are out there. We wanted to put the popular ones to the test and thought Pac-Man is a good benchmark for simple and fast decision making. So we let jev 1.13, kev, clef, clef… - [Bluesky · Pingoru.io · 2026-10-29](https://bsky.app/profile/pingoru.bsky.social/post/3myz5a3dewk2z): 🟠 Berget AI is reporting a Partial Outage since 11:10 UTC "Cloudflare/clef-flash is down" Affects: Cloudflare/clef-flash Live timeline → https://pingoru.io/providers/berget-ai/incidents/12447236 \#BergetAI \#BergetAIDown - [Hacker News · coloneltcb · 2026-10-09](https://news.ycombinator.com/item?id=50025236): Clef-Omni: full multimodality, a faster Clef and a cheaper Clef-flash - [Bluesky · Pingoru.io · 2026-10-27](https://bsky.app/profile/pingoru.bsky.social/post/3myuqamamwa22): 🟠 Berget AI is reporting a Partial Outage since 17:05 UTC "Cloudflare/clef-flash is down" Affects: Cloudflare/clef-flash Live timeline → https://pingoru.io/providers/berget-ai/incidents/12443721 \#BergetAI \#BergetAIDown - [Hacker News · DerDerDaIst · 2026-10-05](https://news.ycombinator.com/item?id=49963695): Clef (and Clef Flash) from Cloudflare - [Bluesky · Pingoru.io · 2026-10-23](https://bsky.app/profile/pingoru.bsky.social/post/3mykelz2iet2b): 🟠 Berget AI is reporting a Partial Outage since 14:10 UTC "Cloudflare/clef-flash is down" Affects: Cloudflare/clef-flash Live timeline → https://pingoru.io/providers/berget-ai/incidents/12436703 \#BergetAI \#BergetAIDown - [Hacker News · joerebil · 2026-10-29](https://news.ycombinator.com/item?id=50293960): Agumon, a 2B model as an inline WAF: Hi HN, Giovanni here, I have been working for the last 10+ years in Cybersecurity, defense operations. While building my cloud platform that's bootstrapping in Europe, I kept reading the access logs of my projects, and wondering over and over how to avoid placing a traditional rule-based WAF as a… - [Bluesky · asura · 2026-10-12](https://bsky.app/profile/asura.dev/post/3mxozs3m75d26): Cloudflare got you huggingface.co/Cloudflare/c... - [Hacker News · nacs · 2026-10-09](https://news.ycombinator.com/item?id=50027543): Typesafe AI raises $870M at $7.5B: Hm? Jev's actual (network) latency is not that great and even small local models are doing far better latency-wise. The Microsoft article GP linked even shows the MS model having 95ms latency. Even on price Jev being matched (the same MS model is "Input tokens cost $0.042 USD per million tokens.… - [Bluesky · Le Fil IA · 2026-10-12](https://bsky.app/profile/le-fil-ia.bsky.social/post/3mxnvq5m6vx22): 1/ Cloudflare sort Clef-omni, modèle de décision multimodal à poids ouverts, huit jours après Clef et Clef-flash. 30 milliards de paramètres, texte, image, audio et vidéo en un appel. Les petits modèles juges deviennent de l'infra. https://www.lefilia.fr/article/6310137 - [Hacker News · murderszn · 2026-10-03](https://news.ycombinator.com/item?id=49947691): Cube Rule food classifier using Cloudflare Clef classification model: I built a small demo that classifies food using the Cube Rule rather than asking a general model to write an essay on whether a hot dog is a sandwich. The Cube Rule categorizes food by the location of structural starch relative to the six faces of a cube: 0 Starchless (Salad): Steak, soup 1 Face… - [Bluesky · NeuroLink · 2026-10-31](https://bsky.app/profile/neurolinkai.bsky.social/post/3mz5es6pgel2t): Cloudflare Clef, 29 Oct probes: 413/5021 fired on an estimate (request JSON, chars ÷ 4, rounded) over twice the window its message prints. Refused: clef-flash 196,609 got 422 (real tokens \> max), 196,610 got 413; clef 524,289 got 422, 524,290 got 413. Mostly-'a'; images not in estimate (24 Oct) ## Trust facts Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions. - Last change on Hugging Face: 2026-11-10 - First published: 2026-09-30 - Checked on Hugging Face: 2026-11-12 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open Cloudflare/clef-flash on Hugging Face](https://huggingface.co/Cloudflare/clef-flash) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/614c30ba-f6b2-46dd-bb0e-4bad233d1941) - [Its page on Mindket](https://mindket.com/p/614c30ba-f6b2-46dd-bb0e-4bad233d1941)