cd /news/artificial-intelligence/inkling-model-from-thinking-machines… · home topics artificial-intelligence article
[ARTICLE · art-60969] src=databricks.com ↗ pub= topic=artificial-intelligence verified=true sentiment=↑ positive

Inkling model from Thinking Machines Lab now on Databricks

Databricks announced that Thinking Machines Lab's open-weights model, Inkling, is now available on its platform through Unity AI Gateway, enabling enterprise customers to build agents and applications with coding and agentic reasoning workflows. The model supports multi-modal inputs and can be fine-tuned on proprietary data, offering enterprises control, choice, and cost optimization.

read2 min views1 publishedJul 15, 2026
Inkling model from Thinking Machines Lab now on Databricks
Image: Databricks Blog

The open-weights model from Thinking Machines Lab is available through the Unity AI Gateway, to build agents and applications

by Mike Eastham, Yuchen Jin and Preslav Le We are excited to announce Databricks as a day zero launch partner for Thinking Machines Lab (TML), bringing their first open-weights model, Inkling, available on the Databricks platform for our enterprise customers to apply to their enterprise data and power coding workflows. Inkling is an open-weights model that excels at coding and agentic reasoning workflows, and supports multi-modal inputs.

Open weight models are core to Databricks’ mission, as they can be customized by our customers on their own enterprise data, to perform specialized tasks with higher quality, lower cost, and often faster latency. Inkling from Thinking Machines Lab (TML), an AI research and products company, continues the recent surge in strong open weight models.

Why this matters for enterprise teams:

Context: Open weight models like Inkling can be fine-tuned on proprietary codebases, internal documentation, and domain-specific data to achieve higher accuracy on your specific tasks.

Control: Inkling is governed through Unity AI Gateway with the same security, permissions, audit logging, and policy enforcement that enterprises apply to all models on Databricks. Data stays within your governed environment.

Choice: No lock-in to a single model provider. Teams can switch, combine, or customize models as requirements evolve and choose the right model for each workload across open and proprietary options.

Cost: Deploy Inkling at the scale and configuration that matches your workloads. Open weight models allow organizations to optimize inference spend without per-token API pricing.

Inkling is available in Unity AI Gateway, our unified governance layer for security, cost controls, and observability. Invoke Inkling via REST API:

Support to query in SQL is coming soon.

Connect Inkling with popular coding agents such as Cursor or OpenCode or Pi through the Unity AI Gateway, and be able to centrally govern access, cost controls and budgets, and security.

Inkling is available now on Databricks. Here's how to get started:

Subscribe to our blog and get the latest posts delivered to your inbox.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @databricks 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/inkling-model-from-t…] indexed:0 read:2min 2026-07-15 ·