\n
\u2192 Scroll horizontally to see all columns<\/p>\n
\n\n\n| Workflow<\/th>\n | Metric<\/th>\n | Clef<\/th>\n | Clef-flash<\/th>\n | Jev<\/th>\n<\/tr>\n<\/thead>\n |
\n\n| Invoice processing<\/td>\n | Exact actions<\/td>\n | 64.7<\/td>\n | 57.1<\/td>\n | 61.8<\/td>\n<\/tr>\n |
\n| Invoice processing<\/td>\n | Primary action<\/td>\n | 86.2<\/td>\n | 73.3<\/td>\n | 83.1<\/td>\n<\/tr>\n |
\n| Customer service<\/td>\n | Exact actions<\/td>\n | 76.3<\/td>\n | 77.0<\/td>\n | 76.0<\/td>\n<\/tr>\n |
\n| Security incidents<\/td>\n | Exact actions<\/td>\n | 62.9<\/td>\n | 61.7<\/td>\n | 61.7<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n In practical workflows such as invoice processing and security incident response, Clef demonstrates higher accuracy than the existing Jev. It records a particularly high figure of 86.2% in identifying primary actions for invoice processing.<\/p>\n Additionally, the performance of the base model Qwen3.8-27B itself is extremely high; in publisher benchmarks, it records 61.7% on SWE-bench Pro (real repository fixes) and 90.3% on LiveCodeBench v6 (competitive programming), possessing performance that overwhelms models of the same scale in coding and agent tasks. Clef is a model built upon this powerful foundation with specialized training for structured outputs.<\/p>\n Strengths and Use Cases<\/h2>\nClef excels at decision-making tasks where it computes and returns probabilities for predefined choices based on a given input “state” and typed “question schemas”. Because it can take not only text and JSON data but also images and video frames simultaneously as state, it enables classification and judgment based on multimodal information.<\/p>\n Supported formats in the question schema include the noul<\/code> type for boolean judgments, the choice<\/code> type for selecting from multiple options, and the score<\/code> type for graded evaluations. Inside the model, a small transformer head (joint schema head) reads the final hidden states of the backbone Qwen3.8-27B (including the vision encoder), and calculates scores (log-probabilities) for all question choices in a single forward pass.<\/p>\nThis mechanism eliminates the need for free-form text generation or complex parsing processes to convert output results into JSON or the like, which were traditionally required with LLMs. Specifically, the following practical use cases are anticipated:<\/p>\n \n- Automatic Invoice and Receipt Judgment<\/strong>: Evaluates payment statuses and amount conditions from image or text-format invoice data.<\/li>\n
- Customer Support Routing<\/strong>: Determines department assignments (billing, technical support, etc.) and urgency (same-day response, within this week, etc.) based on the body of user inquiries.<\/li>\n
- Security Incident Classification<\/strong>: Automatically identifies the presence or absence of service outages from incident reports and log messages.<\/li>\n<\/ul>\n
Furthermore, because it has full compatibility with the APIs of existing decision systems Jev and SystemOne (POST \/v1\/systemone<\/code>), it can be integrated directly into existing workflows that support these formats.<\/p>\nHow It Differs from Similar Models<\/h2>\nCompared to Qwen3.8-27B derivative models previously featured on this site, Clef adopts a clearly distinct approach.<\/p>\n OrcaSAQ-2-27B Text Generation Model: Example Responses and Required VRAM 16GB+<\/a> is a text model compressed to an average of 3.21-bit using proprietary quantization technology (SAQ2) with the objective of running the original 54GB model on a single 16GB GPU. In contrast, Clef does not compress parameters for lightweighting; instead, it retains the vision encoder in the backbone, adds a structured decision head, and supports multimodal inputs.<\/p>\nAdditionally, Hemmingway-1 Everyday Text Language Model: Required VRAM 12GB+ \/ GGUF Available<\/a> is a model specialized in directly generating human-written everyday texts such as emails and messages. Clef, on the other hand, does not generate natural language text; it outputs only probability values (probability distributions) for predefined question choices. The major difference is that while Hemmingway-1 aims to create text, Clef specializes in automating classification and decision-making.<\/p>\n<\/p>\n Hardware Requirements<\/h2>\nEstimated requirements (calculated by Local Model Watch)<\/strong> \u2014 27.4B parameters<\/p>\n |