Inkling Models — Hardware Requirements

3 Inkling models from Thinking Machines and the community, from the smallest that runs in 1.5 GB of VRAM at Q4 up to 952.4B parameters. Every row links to full quantization tables and GPU compatibility.

All Inkling Models by Size

How Inkling Compares — llmrun Composite Score

Inkling Small is the highest-scoring Inkling model with an llmrun Composite Score of 119.6 — #16 among 106 open models. The top proprietary model, Gemini 4 Carbon, scores 158.6. Click a model to see its full benchmark breakdown.

Frequently Asked Questions

How much VRAM do I need to run a Inkling model?

The smallest Inkling model, Inkling Small DSpark Preview, runs from 1.5 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 0.9 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.

Which Inkling models can I run on a 16 GB GPU?

1 of 3 Inkling models fit in 16 GB of VRAM at Q4_K_M, including Inkling Small DSpark Preview.

What is the most popular Inkling model to run locally?

Inkling Small is the most downloaded Inkling model in local-friendly quantized formats. It runs from 162.7 GB of VRAM at Q4_K_M.

How do Inkling models score on benchmarks?

Inkling Small leads the family with an llmrun Composite Score of 119.6, ranking #16 among 106 open models, while the top proprietary model, Gemini 4 Carbon, scores 158.6. See the comparison chart above for the full standings.