Inkling 2 Small overview
Inkling 2 Small by Thinking Machines
Thinking Machines Lab built Inkling 2 Small as a general-purpose mixture-of-experts model with 594 billion total and 17 billion active parameters. It is the first model of the lab's second generation.
Inkling 2 Small specs
- Model ID
thinkingmachines/inkling-2-small- Developer
- Thinking Machines
- Context window
- 1,048,576 tokens
- Max output
- 943,718 tokens
- Input
- Image, Text
- Output
- Text
- API
- OpenAI-compatible, base URL
https://api.aimlapi.com/v1
Where the model is aimed
Thinking Machines Lab recommends Inkling 2 Small for agentic coding, vision understanding and general agentic knowledge work. The model card describes it as intended for natural-language and visual tasks, including coding and tool-use systems.
Only 17 billion of the parameters are active for a given token. The 1M token context window and the output limit of 943,718 tokens leave room for long repositories, long documents and many image inputs in one request.
Tune reasoning effort and sampling
Reasoning effort sets how much the model reasons before it answers. This endpoint accepts minimal, low, medium, high or max. Lower values give faster responses and use fewer tokens on reasoning. Start with medium, then move down for simple extraction and up for multi-step coding or tool-use tasks.
The min p option takes a value from 0.001 to 0.999 and works as an alternative to top p and top k. Use it when you want to cut unlikely tokens without setting the other two samplers. Change one sampling control at a time so you can see its effect.
Check the output on your own tasks
Inkling 2 Small supports streaming, structured output and tools. Run a sample of your real agent tasks at two reasoning effort levels and compare the answers. Validate structured output against your schema and check each tool call before you scale.
Use cases
Run an agentic coding loop that reads a large codebase, calls tools and proposes patches.
Answer questions about screenshots, charts or photos sent together with text.
Handle knowledge work agents that search, call tools and return structured output.
Process long documents in one request using the 1M token context window.
Capabilities
- Reasoning
- Streaming
- Structured outputs
- Function calling
- Vision
Inkling 2 Small API pricing
| Usage | Price | Unit |
|---|---|---|
| Input | $0.3174 | 1M tokens |
| Output | $1.06 | 1M tokens |
| Cached input | $0.0317 | 1M tokens |
Prices from the AI/ML API price list, checked on Nov 13, 2026.
Inkling 2 Small vs other models: token prices
AI/ML API input and output rates for 5 models, per million tokens.
| Model | Input | Output |
|---|---|---|
 Inkling 2 Small | $0.3174 | $1.06 |
| Weaver (alpha) | $0.7935 | $1.06 |
| LFM2.5-24B-A2B | $0.1058 | $1.06 |
| Inkling Small | $0.6136 | $1.52 |
| Inkling | $1.06 | $4.28 |
How to use the Inkling 2 Small API
View docs
Build with your agent
Give your agent the model details, code and documentation.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v1/chat/completions' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "thinkingmachines/inkling-2-small",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "thinkingmachines/inkling-2-small",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}
response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "thinkingmachines/inkling-2-small",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());Models similar to Inkling 2 Small
Inkling 2 Small FAQ
How much does Inkling 2 Small cost?
Inkling 2 Small costs $0.3174 / 1M tokens for input and $1.058 / 1M tokens for output on AI/ML API. Cached input is billed at $0.03174 / 1M tokens.
What inputs does Inkling 2 Small accept?
It accepts text and image input and returns text. It does not generate images.
How do I make responses faster?
Lower the reasoning effort. Lower values give faster responses and use fewer tokens on reasoning.
Which sampling option does this endpoint add?
Min p accepts values from 0.001 to 0.999. It is an alternative to top p and top k.
Does it support tools and structured output?
Yes. Tools, structured output, streaming and vision are all listed capabilities of this endpoint.
What is the Inkling 2 Small model ID?
Use thinkingmachines/inkling-2-small as the model parameter. Requests go to the OpenAI-compatible endpoint with your AI/ML API key.
What is the context window of Inkling 2 Small?
Inkling 2 Small accepts up to 1,048,576 tokens of context and returns up to 943,718 output tokens.