Build customizable enterprise AI systems on an open-weight base model instead of relying entirely on a closed frontier model.
Organizations can start with an open-weight, multimodal model and adapt it to their own needs through fine-tuning, retrieval, behavioral controls, agent configuration, and feedback. The differentiator is the combination of a capable open base and an automated fine-tuning workflow rather than leaderboard performance alone.
online B2B Deep Tech / Infrastructure
From IBM Technology — Thinking Machines Lab drops Inkling & Meta’s Muse Spark 1.1 at 04:44
Problem: A general-purpose closed model may not provide enough control, customization, reproducibility, or task-specific behavior for an organization's needs.
For: Enterprises and agent builders that need AI systems customized for their own data, behavior, workflows, or applications.
Examples
- Thinking Machines' Inkling: a 975 B total-parameter mixture-of-experts model with 41 B active parameters, presented as an open-weight base intended for customization even though it was not the strongest available model.
Behind this: 12 build steps · 2 tools and how each is used · how to validate demand · 2 more real examples · 6 things the video never answers.
Other takes on AI model development
- Retrain small open-source language models for long-horizon agentic workloads and sell efficient inference and on-device deployments
- AI simulation company that models human behavior and lets organizations test decisions against simulated populations
- An open foundation-model company focused on coding and long-horizon software tasks
- Enterprise context and training-data infrastructure for AI applications
- In-database semantic analytics using large database models
- A domain-specific AI product built around private data, specialized models, and purpose-built workflows