{"slug": "simple-jev-by-featherless-ai-turn-context-into-clear-decisions", "title": "Simple-jev by featherless AI – Turn context into clear decisions", "summary": "Featherless AI released Simple-jev, an open-source tool that turns existing open language models into classifiers by reading next-token logits for allowed labels, returning structured JSON with choices, confidence scores, and probabilities. The demo API runs the featherless-ai/gemma-4-26B-A4B-classifier model with limits of 2,000 tokens and 4 requests per second, and image context is supported only for Gemma and Qwen models. Simple Jev models on Featherless start at $0.03 per million input tokens, with the service currently in beta and hosted fine-tuned models and fine-tuning planned as usage grows.", "body_md": "### Bring the whole conversation.\n\n              We support chat-format context. Send a conversation as\n              `messages` with roles, then ask questions about the\n              exchange—not just the last message.\n            \n\n[Explore chat context](docs.html#context)\n\nOPEN MODELS. STRUCTURED DECISIONS.\n\nTurn context into clear decisions.\n\n            Give __any__ open model a message and a few questions.\n\n            Get choices, scores, and answers your code can use.\n            \n\n[Source on Github here](https://github.com/featherless-ai/simple-jev) \n\nA little model.\n\nA very specific job.\n\n```\ncurl https://simple-jev-demo-api.featherless.ai/v1/classifier \\\n  -H 'Content-Type: application/json' \\\n  -d '{\n    \"model\": \"featherless-ai/gemma-4-26B-A4B-classifier\",\n    \"state\": \"Mia owns a red bicycle.\",\n    \"questions\": {\n      \"color\": {\n        \"type\": \"choice\",\n        \"instructions\": \"What color is Mia’s bicycle?\",\n        \"criteria\": {\"red\": null, \"blue\": null}\n      }\n    }\n  }'\n\n## For production usage and higher rate limits,\n## please signup for a developer account at https://featherless.ai\n{\n  \"color\": {\n    \"type\": \"choice\",\n    \"choice\": \"red\",\n    \"confidence\": 1,\n    \"probabilities\": {\n      \"red\": 1,\n      \"blue\": 1.8874485308018052e-10\n    }\n  }\n}\n```\n\n          Actual Gemma response, showing the `answers` field. Results\n          may vary. Demo limits: 2k tokens · 4 requests/second.\n          [Full API documentation ↗](docs.html)\n\n01 / TRY IT\n\nEdit scenarios, build questions, and see real model responses in the interactive playground.\n\nSEE IT IN ACTION\n\nMORE WAYS TO BRING CONTEXT\n\n              Use images as context for your questions. Image and vision support\n              is available for **Gemma and Qwen models only**.\n            \n\n02 / THE IDEA\n\n            Use existing open language models as classifiers.\n\nNo separate\n            classifier head required.\n          \n\nA message, chat history, or structured state. Ask several questions about the same input.\n\nWe read the next-token logits for the allowed labels and reuse the shared prompt cache across questions.\n\nYour application receives structured JSON. Route a ticket, rank a result, or choose the next step.\n\nONE MORE THING\n\n**Really Fancy Decision Training.** Bring examples\n              and answers—or ask a larger teacher model to label them. RFDT\n              fine-tunes the decisions your application actually needs.\n            \n\nREADY FOR MORE?\n\n            Simple Jev models on Featherless start at\n            **$0.03 per million input tokens**, with higher limits\n            on developer plans. Currently in beta.\n\nSupport for hosted\n            fine-tuned models and fine-tuning is on the way as usage grows.\n          \n\n*Prices may change after beta. Refer to\n              [Featherless.ai official pricing](https://featherless.ai/)\n              for up-to-date information.", "url": "https://wpnews.pro/news/simple-jev-by-featherless-ai-turn-context-into-clear-decisions", "canonical_source": "https://simple-jev.featherless.ai/", "published_at": "2026-09-22 22:34:36+00:00", "updated_at": "2026-09-22 22:54:35.166014+00:00", "lang": "en", "topics": ["ai-tools", "large-language-models", "ai-products", "developer-tools", "ai-infrastructure"], "entities": ["Featherless AI", "Simple-jev", "Gemma", "Qwen", "featherless-ai/gemma-4-26B-A4B-classifier", "RFDT"], "alternates": {"html": "https://wpnews.pro/news/simple-jev-by-featherless-ai-turn-context-into-clear-decisions", "markdown": "https://wpnews.pro/news/simple-jev-by-featherless-ai-turn-context-into-clear-decisions.md", "text": "https://wpnews.pro/news/simple-jev-by-featherless-ai-turn-context-into-clear-decisions.txt", "jsonld": "https://wpnews.pro/news/simple-jev-by-featherless-ai-turn-context-into-clear-decisions.jsonld"}}