# Baseten — Introducing Inkling-Small: small and mighty multimodal AI

- Company: Baseten (baseten.co)
- Announced: 2026-09-23
- Category: availability-change
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://www.baseten.co/blog/introducing-inkling-small
- Record: https://forck.live/items/34734-introducing-inkling-small-small-and-mighty-multimodal-ai
- Subject: Baseten
- Open weights: yes
- Models affected: Inkling-Small, Inkling
- Context window: 1M token context window

Baseten has added Thinking Machines' Inkling-Small to its Model APIs and Dedicated Inference, giving customers a smaller, cheaper way to run the same multimodal capabilities as Inkling. The model keeps Inkling's text, image and audio inputs and its 1M token context window, so Baseten describes it as an ideal drop-in replacement. At roughly one-quarter the size of Inkling, it uses a sparse Mixture-of-Experts architecture that lowers VRAM requirements, which reduces deployment costs, allows use on fewer GPUs and leaves headroom for faster inference. Baseten says Inkling-Small delivers comparable performance to Inkling across reasoning, coding, vision, audio and safety, and slightly outperforms it on several benchmarks, though individual scores fluctuate. Like Inkling, it is open-weight, and Baseten runs it on NVIDIA accelerated infrastructure. Baseten says its Delivery Network gives 2–3× faster cold starts, and that multi-cloud capacity management keeps applications online when specific cloud resources are constrained.

## Evidence

Verbatim from https://www.baseten.co/blog/introducing-inkling-small:

> Thinking Machines’ new Inkling-Small is now available on Baseten via Model APIs and Dedicated Inference. It retains the powerful multimodal capabilities and 1M-token context window of the original Inkling, but with a significantly smaller footprint that delivers reduced deployment costs and faster inference.
> 
> Thinking Machines' Inkling-Small is now available on Baseten through Model APIs and Dedicated Inference , and it runs on NVIDIA accelerated infrastructure.
> 
> Earlier this month, we launched Inkling on Baseten . Inkling is a highly knowledgeable, open-weight multimodal foundation model that combines broad reasoning capabilities, long context, and efficient sparse inference, making it a flexible base for customized AI applications.
> 
> Inkling-Small retains Inkling’s compelling model characteristics, while providing faster inference and lower cost due to its smaller footprint.
> 
> A smaller footprint
> 
> At roughly one-quarter the size of Inkling, Inkling-Small leverages a sparse Mixture-of-Experts (MoE) architecture to deliver exceptional efficiency.

---

Record: https://forck.live/items/34734-introducing-inkling-small-small-and-mighty-multimodal-ai
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md
