TechCompare LogoTechCompare

Inkling API Pricing & Cost Calculator

By TechCompare · Updated · How we check

Inkling suits buyers who want a model from outside OpenAI, Anthropic, Google, and the Chinese labs. It lists below GPT-5.6 Terra on both rows, with a 1M context for high-volume document work.

Calculator

Cost Comparison

Based on 50,000 input tokens (50% cached), 3,000 output tokens, and 500 requests.Prices come live from OpenRouter at each maker's own price where it's listed. Temporary deals are left out.

Loading model prices
Same 50,000 in / 3,000 out (50% cached) scenario, pre-loaded. Adjust tokens, caching, or swap models.
Edit in the full tool

Inkling is Thinking Machines Lab's model and their first broadly available API release. It's priced at or below GPT-5.6 Terra, with a 1-million-token context for agentic workloads.

Thinking Machines: Inkling

Cheapest host: Together

Input
$1
Output
$4.05
Cached
$0.17

per 1M tokens

Price checked 2026-11-12

How this is calculated

Inkling is priced at $1.00 per million input tokens and $4.05 per million output tokens, with a prompt caching discount of 83% ($0.17 per million cached reads). That's 50% below GPT-5.6 Terra's input list rate ($2.00), while output sits below GLM-5.2 ($4.40) and below Terra.

Verdict

Choose Inkling when you want a model from outside the OpenAI, Anthropic, Google and Chinese lab stacks, since its input rate stays below GPT-5.6 Terra's and its output rate below GLM-5.2's.

Inkling's $1/M input and $4.05/M output sit 50% below GPT-5.6 Terra's $2 input rate, with cached input at $0.17/M (83% discount). Output at $4.05/M undercuts GLM-5.2's $4.40 but sits above Gemini 3.6 Flash's $3.75. The 1M context window fits the same document-heavy work as those peers.

More API Standalones scenarios

Claude Fable 5.1 vs GPT-6 Astra
Compare sticker prices on both tiers and see how much the cache discount changes the bill.
View details
Claude Fable 5 vs Claude Fable 5.1
Same generation. Compare sticker prices and cache discounts to see what 5.1 changes.
View details
Claude Haiku 5.5 vs GPT-6 Luna
Small fast models at volume, with cache reads, batch jobs, and Haiku's long prompt tier in the math.
View details

Related guides

Inkling 975B MoE VRAM estimate
VRAM footprint for the Inkling model this pricing entry is for.
View guide
Token cost calculator
How per-token counts turn into API costs at Inkling's $1/MM rates.
View guide

Frequently asked questions

Who is Thinking Machines Lab and why is Inkling notable?
Thinking Machines Lab is a recent AI startup, and Inkling is their first broadly available model on OpenRouter. It's a mid-tier alternative to GPT-5.6 Terra and GLM-5.2 with a 1M context window.
How much does Inkling cost per million output tokens?
Inkling charges $4.05 per million output tokens at standard pricing, with input at $1.00 per million. The 83% prompt caching discount brings cached input to $0.17 per million, and no batch discount is currently advertised.

Related tools

LLM VRAM Calculator

Calculate the VRAM needed to run or fine-tune any LLM at any quantization.

Use tool

Power Cost Estimator

Estimate annual electricity costs for your PC, Server, or TV.

Use tool

Data Transfer Calculator

Estimate transfer times for files over USB, WiFi, Ethernet, and more.

Use tool

JSON Formatter

Validate, format, and minify JSON. Browse it as a tree, run JSONPath queries, compare two documents, or validate against a JSON Schema.

Use tool