Inkling vs Inkling Small.

Thinking Machines' two open models: full-size depth against a faster, cheaper sibling.

Inkling and Inkling Small are the same idea at two sizes. Both read images, support tools and offer zero retention; the full Inkling is built for general reasoning and coding, while Inkling Small is tuned for quick multimodal answers.

Inkling has 975B parameters, a million-token window and a $4.05 price. Inkling Small has 276B parameters with 12B active per token, a 524k window and a $1.20 price. The smaller model gives up context and scale, not capabilities.

At a glance

Where Inkling pulls ahead

• The full-size model at 975B parameters for harder general reasoning and coding.
• A million-token window, nearly twice the 524k of Inkling Small.
• Flagship-level depth when the task justifies the larger model.

Where Inkling Small pulls ahead

• Price: $1.20 versus $4.05 per million tokens, about 70% cheaper.
• Only 12B active parameters per token, so it answers much faster.
• The same image input, tool support and zero-retention terms at a much lower cost.

The price difference

Inkling costs $4.05 per million tokens against $1.20 for Inkling Small — about 3.4x more. On UsingOpen you can chat with both for free inside the daily limit, so the rate is a fair proxy for how expensive each model is to run rather than a bill you pay here. Multiply it by your monthly token volume if you are comparing API access elsewhere.

Which should you choose?

Inkling Small is the sensible default for quick multimodal questions and everyday coding help. It keeps the capabilities that matter and cuts the price by more than two thirds.

Reach for the full Inkling when the problem is hard enough to need 975B parameters or the extra context headroom. For most conversations, the small model will answer faster and nearly as well.

Chat with Inkling free · Chat with Inkling Small free

Frequently asked questions

Which has the larger context window, Inkling or Inkling Small?

Inkling is listed at 1M (750,000 words); Inkling Small is listed at 524k (390,000 words). Inkling can hold more in one conversation, which matters for long codebases, archives and multi-hour agent runs.

Can both Inkling and Inkling Small read images?

Yes. Both models accept image input, so a screenshot or diagram is fair game in either. The decision should turn on context window, price, retention terms and how each model answers your own prompts.

Is Inkling or Inkling Small better for coding?

Inkling is the closer fit for code here — the catalog tunes it for general reasoning, coding and tools — while Inkling Small is aimed at quick multimodal answers. For project-scale coding, start with Inkling; if the task leans toward the other model's strengths, it may still win.

Which is cheaper, Inkling or Inkling Small?

Inkling Small is listed at $1.20 per million tokens against $4.05 for Inkling — about 3.4x cheaper. Both are free to chat with on UsingOpen inside the daily limit when live; the rates matter if you are comparing what API access would cost.

Can I chat with Inkling and Inkling Small for free?

Yes. Both are selectable on UsingOpen without an account when they are live, and each user gets 10 free prompts per day. A model paused for capacity is unavailable until it returns; the picker shows live status. History stays on your device, and Compare mode can send one prompt to both models side by side.

How much context does Inkling Small have?

Inkling Small is listed with a 524k-token window, roughly 390,000 words. The full Inkling doubles that to one million tokens, about 750,000 words.

Is the full Inkling worth 3.4 times the price?

Only for tasks where the extra scale or context shows a measurable gain. Both models read images, support tools and offer zero retention. Test the full model on your hardest prompt; if the answers are close, keep Inkling Small.

Related comparisons

• Qwen3.8 2.4T vs Kimi K3 — The two trillion-parameter flagships: Qwen's text heavyweight against Kimi's multimodal giant.
• MiMo V2.6 Pro vs Qwen3.8 Flash — Xiaomi's flagship against Alibaba's cheaper fast multimodal model.
• Muse Glimmer 30B vs Gemma 4 31B — Meta's compact agent model against Google's cheaper writer and image reader.

All models · Compare models · Free AI chat · FAQ