Announcing Together AI and Adaption Partnership

2026-04-30 · Source: Together AI | The AI Native Cloud - Together.ai · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Intermediate, quick

Summary

Together AI and Adaption have partnered to integrate Together Fine-Tuning directly into Adaption's Adaptive Data platform, launched on April 30, 2026. This collaboration allows users to optimize training datasets within Adaptive Data, which is co-founded by former Cohere and Google DeepMind leaders Sara Hooker and Sudip Roy and reports an average 82% increase in data quality. Following data optimization, users can seamlessly execute Together Fine-Tuning with optimized hyperparameters. Together AI's platform supports fine-tuning leading open models, including those over 100B parameters like Kimi K2.5, GLM 5.1, and Qwen 3.5-397B, for structured tool use, reasoning, and vision-language setups. The integration streamlines the workflow from data preparation to model deployment on Together AI's high-performance inference service, offering faster development of high-quality, fine-tuned open models with experiment visibility.

Key takeaway

For AI Engineers or ML teams focused on customizing open models, this partnership offers a streamlined workflow to achieve higher quality fine-tuned models faster. If you are struggling with data preparation bottlenecks or inconsistent fine-tuning results, consider using Adaptive Data to optimize your datasets. This integration allows you to directly execute fine-tuning on Together AI's infrastructure, reducing manual steps and improving model performance against target behaviors.

Key insights

Optimizing training data before fine-tuning open models significantly enhances model quality and development speed.

Principles

• Data quality directly impacts fine-tuned model performance.
• Integrated platforms streamline complex ML workflows.
• Experiment visibility accelerates model adaptation.

Method

Users optimize training data in Adaptive Data, then execute Together Fine-Tuning with optimized hyperparameters, evaluate results, and deploy the model on Together AI's inference service.

In practice

• Use Adaptive Data to analyze and adapt datasets.
• Fine-tune large open models like Kimi K2.5.
• Monitor win rates and loss during fine-tuning.

Topics

• Together AI
• Adaption
• Fine-tuning
• Adaptive Data
• Open Models
• Data Quality
• Model Deployment

Best for: MLOps Engineer, NLP Engineer, Computer Vision Engineer, Machine Learning Engineer, AI Engineer, Director of AI/ML

Related on AIssential

• Choosing the Right Model is Hard. Maintaining Accuracy is Harder. — Continuous monitoring and iterative refinement of open-source LLMs in production can enhance accuracy and efficiency.
• TeamTTA: Efficient Multi-Device Collaboration for Open-Set Test-Time Adaptation via Cloud Integration — Cloud-edge collaboration with LVM-driven knowledge distillation improves edge DNN adaptation in open-set, dynamic environments.
• ADAPTive Input Training for Many-to-One Pre-Training on Time-Series Classification — ADAPT enables mixed-batch pre-training for time-series data by aligning physical properties across diverse datasets.
• Adaptive ML boss: To win in AI, businesses must build with open source — Open-source AI models offer businesses a strategic advantage over proprietary solutions.
• On the slow death of Scaling (birth of Adaption Labs) | Sara Hooker | HF ML Club India EP2 — Monolithic AI scaling is ending; the future is adaptive, efficient, post-training intelligence.
• What properties of reasoning supervision are associated with improved downstream model quality? — Intrinsic data metrics can reliably predict reasoning dataset utility before model training, with scale-dependent predictors.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Together AI | The AI Native Cloud - Together.ai.