https://finetunedb.com [fdb-square]FinetuneDB Features Security Docs Use cases Case studies Blog --------------------------------------------------------------------- Log in Get started for free [fdb-square] FinetuneDB Features Security Docs Use cases Case studies Blog Log in Get started for free [menu] Build fine-tuned AI models with your data AI Fine-tuning Platform to Create Custom LLMs Train AI models with your data in minutes, not weeks, and get better performance at lower cost. Get started for free Book a demo Chat preview Integrates with open-source and proprietary foundation models [open] [jkfg] [kY46x] [1] [60337] [dataset] Manage multiple models and datasets all in one place [datasets] Collaborative editor Build datasets together Collaborate with your team to build a proprietary fine-tuning dataset for your AI models [evals] Datasets Create your AI Moat Build a proprietary fine-tuning dataset to differentiate model performance from your competition [evals] Improve Fine-tuning model versions Build custom fine-tuning datasets to optimize model performance for specific use cases Improve Fine-tuning model versions Build custom fine-tuning datasets to optimize model performance for specific use cases Datasets Create your AI Moat Build a proprietary fine-tuning dataset to differentiate model performance from your competition [chains] Collaborative editor Build datasets together Collaborate with your team to build a proprietary fine-tuning dataset for your AI models [datasets] [evaluation] Evaluate model performance with human and AI feedback [copilot] Copilot Improve outputs automatically Copilot automates evaluations and model improvements for your use case [evals] Evaluators Benchmark outputs Collaborate with product managers, domain experts and engineers to build reliable model outputs [metrics] Metrics Measure model metrics Track AI metrics such as speed, quality scores, and token usage Evaluators Benchmark outputs Collaborate with product managers, domain experts and engineers to build reliable model outputs Metrics Measure model metrics Track AI metrics such as speed, quality scores, and token usage [metrics] Copilot Improve outputs automatically Copilot automates evaluations and model improvements for your use case [copilot] [logs] Collect production data to fine-tune your model [logs] Log view Log production data Capture user interactions, model responses, and system metrics, for ongoing model improvement [filters-mo] Filters Refine your search Use advanced filters to search with precision, so nothing slips past the cracks [chains] Tracing Visualize language chains Trace unlimited nested actions and get a detailed view of the entire request Log view Log production data Capture user interactions, model responses, and system metrics, for ongoing model improvement [logs] Filters Refine your search Use advanced filters to search with precision, so nothing slips past the cracks [filters] Tracing Visualize language chains Trace unlimited nested actions and get a detailed view of the entire request [chains] [studio] Create prompts collaboratively for best performance [studio] Studio Prompt playground Create, manage, and optimize prompts to achieve precise and relevant interactions between users and AI models [evals] Evaluators Measure your models performance Effortlessly track, compare, and refine your prompt versions [evals] Versions Keep track of changes Involve domain experts to manually evaluate outputs, or use automated metrics Studio Prompt playground Create, manage, and optimize prompts to achieve precise and relevant interactions between users and AI models [studio] Versions Keep track of changes Effortlessly track, compare, and refine your prompt versions [filters] Models Compare different models Compare foundation models, and fine-tuned versions to improve prompt performance and save tokens [chains] Developers Easy integration into your production applications // import OpenAI from 'openai' import OpenAI from "finetunedb/openai"; // Fully compatible with original OpenAI initialization const completion = await openai.chat.completions.create({ messages: [{ role: "user", content: "Hello world!" }], model: "gpt-3.5-turbo", // Log requests to FinetuneDB finetunedb: { projectId: "cloio7t90000...", logRequest: true, tags: ["test-prompt"], }, }); Compatible with OpenAI SDK Our SDK extends the OpenAI SDK, allowing you to get started within minutes SDKs for Python & JS/TS Capture production data and send it fully async to FinetuneDB. You have full control on what is sent Web API Need more control? Use the FinetuneDB APl to ingest logs and build your own integrations Langchain integration Get full execution traces in 5 minutes by adding the FinetuneDB callback handler to your app SDKs for Python & JS/TS Typed SDKs that capture production data and send it fully async to FinetuneDB. You have full control on what is sent Web API Need more control? Use the FinetuneDB APl to ingest logs and build your own integrations Langchain integration Get full execution traces in 5 minutes by adding the FinetuneDB callback handler to your app Compatible with OpenAI SDK Our SDK extends the OpenAI SDK, allowing you to get started within minutes // import OpenAI from 'openai' import OpenAI from "finetunedb/openai"; // Fully compatible with original OpenAI initialization const completion = await openai.chat.completions.create({ messages: [{ role: "user", content: "Hello world!" }], model: "gpt-3.5-turbo", // Log requests to FinetuneDB finetunedb: { projectId: "cloio7t90000...", logRequest: true, tags: ["test-prompt"], }, }); [security][security] Hardened Security Your data is end-to-end encrypted Secure encryption of data All data is encrypted in transit using TLS 1.2+ with perfect forward secrecy. Servers holding data use industry-standard AES 256 encryption Strict permissions Manage user roles and permissions. If there are any changes to permissions, FinetuneDB will reflect those changes immediately SOC 2 compliance We're actively working towards achieving SOC 2 compliance to ensure that our systems and processes are secure and reliable Secure encryption of data All data is encrypted in transit using TLS 1.2+ with perfect forward secrecy. Servers holding data use industry-standard AES 256 encryption. Strict permissions enforcement Manage user roles and permissions. If there are any changes to permissions, FinetuneDB will reflect those changes immediately SOC 2 compliance We're actively working towards achieving SOC 2 compliance to ensure that our systems and processes are secure and reliable AI Fine-tuning Platform to Create Custom LLMs Get started for free Book a demo [fdb-square]FinetuneDB * Terms of service * Privacy policy * Cookies * Consent Preferences * [email protected] Made in Stockholm - LLM Ops AB