thinkingmachines · Open weights
Inkling Small Thinking
The reasoning version of Thinking Machines' 276B-parameter open-weights multimodal Mixture-of-Experts model with 12B active parameters. It reasons over text and images, and is designed for agentic coding, tool use, instruction following, and long workflows with controllable thinking effort.
- Reasoning
- Tools
- Vision
- Audio input
- Input
- Text, Images, Audio
- Output
- Text
- Reasoning effort
- minimal · low · medium · high · max
- Discovery tier
- Open weights
Pricing
USD per 1M tokens- Input
- $0.45
- Output
- $1.20
- Cache read
- $0.10
Reference prices as published by the model provider. API example
Specifications
- Context window
- 524.3K
- Max output
- 32.8K
Side by side
Inkling Small Thinking compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Inkling Small ThinkingthinkingmachinesThis model | Inkling ThinkingthinkingmachinesRemove | |
|---|---|---|
| Provider | thinkingmachines | thinkingmachines |
| Discovery tier | Open weights | Open weights |
| Added | Jul 31, 2026 | Jul 15, 2026 |
| Limits | ||
| Context window | 524.29K | 1.04MBest |
| Max output | 32.77K | 32.77K |
| Average output | — | 1.2K tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.45Best | $1.00 |
| Output | $1.20Best | $4.05 |
| Cache read | $0.10Best | $0.17 |
| Benchmark indexes | ||
| Intelligence Index | 25.7Best | 25.0 |
| Coding Index | 52.9Best | 52.1 |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | Yes | Yes |
| Vision | Yes | Yes |
| Structured output | No | Yes |
| Audio input | Yes | Yes |
| Reasoning effort | minimal, low, medium, high, max | minimal, low, medium, high, xhigh |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Inkling Small Thinking
- Inkling Thinking
Intelligence Index
- Inkling Small Thinking25.7
- Inkling Thinking25.0
Inkling Small Thinking: Better than 62% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- Inkling Small Thinking: 52.9Top
- Inkling Thinking: 52.1
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Inkling Small Thinking: 4.8%
- Inkling Thinking: 5.0%Top
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Inkling Small Thinking: 0.6%Top
- Inkling Thinking: 0.3%
Harvey LAB-AAv1.1
Legal tasks passed in full, with no material hallucination
- Inkling Small Thinking: 0.0%Top
- Inkling Thinking: 0.0%Top
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Inkling Small Thinking: 911 EloTop
- Inkling Thinking: 832 Elo
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Inkling Small Thinking: 1,124 EloTop
- Inkling Thinking: 1,072 Elo
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Inkling Small Thinking: 11.2%
- Inkling Thinking: 12.8%Top
AA-LCRv1.1
Reasoning across long documents
- Inkling Small Thinking: 75.7%
- Inkling Thinking: 77.3%Top
MLCR-AA
Long-context reasoning over medical records
- Inkling Small Thinking: 7.8%Top
- Inkling Thinking: 6.1%
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Inkling Small Thinking: 33.3%Top
- Inkling Thinking: 31.9%
GPQA Diamond
Graduate-level science questions
- Inkling Small Thinking: 89.5%Top
- Inkling Thinking: 87.2%
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Inkling Small Thinking: 1.0%Top
- Inkling Thinking: 1.0%Top
SciCode
Scientific computing in Python
- Inkling Small Thinking: 49.7%Top
- Inkling Thinking: 47.0%
Legacy evaluations
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Inkling Small Thinking: 75.7%
- Inkling Thinking: 77.3%Top
Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "thinkingmachines/Inkling-Small:thinking",
"reasoning_effort": "medium",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="thinkingmachines/Inkling-Small:thinking",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)thinkingmachines
Related models
Other thinkingmachines models, starting with the same family.