Thinking Machines Inkling API in India

Inkling is an open-weight model from Thinking Machines Lab, live on unoblox as thinkingmachines/inkling. It takes text and image input, supports reasoning and tool calling, and bills in rupees on a GST invoice.

What Inkling is

Thinking Machines released Inkling on 15 July 2026 as an open-weight model under the Apache 2.0 licence. It is a mixture-of-experts model with about 975 billion total parameters and 41 billion active per token. unoblox lists a context length of 524,288 tokens, with text and image input and text output.

Thinking Machines states a context window of up to 1M tokens for the model; the 524,288-token limit listed on unoblox is the one that applies to your requests.

Where it fits

• Reasoning-heavy assistants that work through multi-step problems.
• Image plus text analysis such as reports with charts.
• Agent tasks with tool use over long sessions.

Try Inkling when you want a recent open-weight model with reasoning, tools and image input. Compare with Qwen3.5 397B A17B and MiMo V2.6 Pro on your own evaluation set, as results vary by task.

₹ pricing for Inkling

On unoblox, Inkling costs ₹91.47 per million input tokens and ₹389.94 per million output tokens. That displayed price is what you are billed per token; nothing is added on top. Cached input tokens are listed at ₹15.41 per million. The model's context length on unoblox is 524,288 tokens.

Model	Input (₹ / 1M tokens)	Output (₹ / 1M tokens)
Inkling	₹91.47	₹389.94
Qwen3.5 397B A17B	₹43.33	₹288.85
MiniMax M3	₹26.96	₹105.91
MiMo V2.6 Pro	₹41.40	₹83.77

As a worked example, a month with 20 million input tokens and 4 million output tokens comes to about ₹3,389.16 on Inkling (₹1,829.40 for input plus ₹1,559.76 for output). Live rates are always on the model page, and each request is rounded up to the nearest paisa.

Calling Inkling from the unoblox endpoint

unoblox is OpenAI-compatible, so any OpenAI SDK works by changing the base URL and key.

curl https://api.unoblox.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_UNOBLOX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "thinkingmachines/inkling",
    "messages": [{"role": "user", "content": "Summarise this support ticket in two lines."}],
    "max_tokens": 512
  }'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_UNOBLOX_API_KEY",
    base_url="https://api.unoblox.ai/v1",
)
resp = client.chat.completions.create(
    model="thinkingmachines/inkling",
    messages=[{"role": "user", "content": "Summarise this support ticket in two lines."}],
    max_tokens=512,
)
print(resp.choices[0].message.content)

For image input, pass an image_url content part:

resp = client.chat.completions.create(
    model="thinkingmachines/inkling",
    messages=[{"role": "user", "content": [
        {"type": "text", "text": "What does this chart show?"},
        {"type": "image_url", "image_url": {"url": "https://example.com/chart.png"}},
    ]}],
    max_tokens=512,
)

Get a key at https://unoblox.ai/sign-in and top up your wallet in rupees.

Notes on parameters, billing and data

Use max_tokens for output limits. Send images as image_url parts. The model runs at an external provider, so we claim no India data residency. Thinking Machines lists audio input for the model, which unoblox does not expose. unoblox provides rupee pricing, GST invoicing and one endpoint.

Frequently asked questions

Is Inkling available in India? Yes, as thinkingmachines/inkling on https://api.unoblox.ai/v1.

Who makes it? Thinking Machines Lab.

Is it open weight? Yes, under Apache 2.0.

What context length does unoblox list? 524,288 tokens.

Does it accept images? Yes, text and image input.

Does it support reasoning and tools? The catalogue lists both.

Get started in rupees → https://unoblox.ai/sign-in