Inkling model from Thinking Machines Lab now on Databricks

2026-07-15 · Source: Databricks · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Cloud Computing & IT Infrastructure · Depth: Intermediate, quick

Summary

Thinking Machines Lab (TML) has partnered with Databricks to launch its first open-weights model, Inkling, on the Databricks platform. Inkling is designed for enterprise customers, excelling in coding and agentic reasoning workflows, and supports multi-modal inputs. This integration allows enterprises to fine-tune Inkling on proprietary data for higher accuracy, maintain control through Unity AI Gateway's security, permissions, and audit logging, and avoid vendor lock-in. Organizations can optimize inference costs without per-token API pricing. Inkling is available via REST API through Unity AI Gateway, with SQL query support coming soon. It can also connect with coding agents like Cursor or OpenCode, and users can get started via AI Playground, Unity AI Gateway deployment, or by building agents with Agent Bricks.

Key takeaway

For AI Engineers or MLOps teams seeking to integrate advanced coding models, Inkling on Databricks offers a governed, customizable solution. You can fine-tune Inkling on your proprietary data to boost accuracy for specific tasks, ensuring data security and cost optimization through Unity AI Gateway. Consider deploying Inkling via the Gateway to centralize access control and connect with your preferred coding agents like Cursor or OpenCode.

Key insights

Inkling, an open-weights coding model, is now on Databricks, offering enterprises customizable, secure, and cost-effective AI for coding workflows.

Principles

• Open-weight models enable customization on proprietary data.
• Unified governance layers ensure enterprise security and control.
• Flexible model choice reduces vendor lock-in and optimizes costs.

Method

Deploy Inkling via Unity AI Gateway, invoke through REST API, or connect with coding agents like Cursor or OpenCode for governed access.

In practice

• Fine-tune Inkling on internal codebases for specific tasks.
• Use Unity AI Gateway for centralized security and cost controls.
• Integrate Inkling with existing coding agents for enhanced workflows.

Topics

• Inkling
• Databricks
• Open-weight Models
• Unity AI Gateway
• Coding Workflows
• Enterprise AI

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Engineer, Machine Learning Engineer, MLOps Engineer

Related on AIssential

• The AI That Can Re-Write Its Own Brain: Why Inkling is the New Frontier for Open Weights — Inkling enables autonomous AI development and dynamic intelligence scaling through an open-weights, self-modifying architecture.
• OpenAI GPT-5.5 + Codex, now available and fully-governed in Databricks — Databricks integrates OpenAI's GPT-5.5 and Codex with Unity AI Gateway for governed enterprise AI applications.
• Stripe data now available on Databricks via Databricks Marketplace — Direct Stripe data integration into Databricks Unity Catalog enables real-time analytics and AI applications.
• not much happened today — Rapid advancements in non-invasive brain-to-text, efficient AI inference, and sophisticated agent orchestration are reshaping AI development.
• Together AI brings Thinking Machines Lab’s new model Inkling on day 0 — Inkling unifies multimodal inputs and offers controllable reasoning via a novel MoE architecture for broad task versatility.
• not much happened today — Inkling's launch emphasizes an open-weight, customizable multimodal foundation model designed for practical application and ecosystem integration over pure benchmark leadership.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Databricks.