Models / 
Thinking Machines Lab

Thinking Machines Lab: the Inkling family of open models

Try open-weight models from Thinking Machines Lab, built for reasoning, agentic coding, and multimodal understanding, served on Crusoe infrastructure.

Benefits of running Thinking Machines Lab on Crusoe Cloud

Start with a model built to be fine-tuned

Thinking Machines Lab designed Inkling to be a starting point, not a finished product, an open model teams adapt to their own data and expertise. On Crusoe, run it on a dedicated endpoint through Self-Serve Deployments, with capacity you control rather than a shared queue.

Pick the precision that fits your budget

Inkling ships in full and quantized builds. Run the quantized version to cut memory use and raise throughput on the same GPU, or the full build when accuracy matters most, each on its own dedicated endpoint through Self-Serve Deployments.

Your data stays on Crusoe Cloud

Inkling runs entirely on Crusoe-operated infrastructure. Your prompts, outputs, and fine-tuning data stay with Crusoe and are never sent to Thinking Machines Lab.

All Thinking Machines Lab models

Text
Vision
Voice
Self-serve
TML

Inkling Small

Context
1048576
Text
Vision
Voice
Self-serve
TML

Inkling Small FP8 Dynamic

Context
1048576

Fine-tune and deploy in Crusoe Intelligence Foundry

Serverless
Inference

Get started fast with usage-based consumption of top open models via a fully managed inference API.

Best for
• Early-stage workloads 
• Low volume traffic
• Rapid experimentation

Self-Serve Deployments

Serve models with a dedicated endpoint, optimized for throughput or responsiveness.

Best for
• Production-scale workloads
• Low-latency applications
• Predictable costs

Tailored
Deployments

Work with our team for the highest level of optimization and a dedicated, benchmarked endpoint. You bring the model and define the requirements, we handle the rest.

Best for
• Proprietary models
• Bespoke inference optimization
• SLA-backed performance

Serverless
Fine-Tuning

Quickly adapt models to your domain. You bring the dataset and pick a base model. We handle the GPUs, the training job and the checkpoints. Go live with Self-Serve Deployments in one click, or download your weights to deploy anywhere.

Best for
• Domain-specific accuracy
• Consistent output formatting
• Smaller models, lower cost

Frequently
asked questions

Are you ready to build something amazing?