Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Text
  3. Tee

Inkling 2 Small TEE

Inkling 2 Small is Thinking Machines Lab's open-weight multimodal Mixture-of-Experts model, with 594B total and 17B active parameters, released under Apache 2.0. It accepts text and images and is built for agentic coding, tool use, visual reasoning and general chat, with adjustable reasoning effort and a 524,288-token context window shared by input and output. Running inside a TEE (Trusted Execution Environment) with attestation support for more private, verifiable inference. This route is experimental, so availability may change.

  • Reasoning
  • Vision
  • Tool Calling
  • Parallel Tool Calls
  • Structured Output

Added Nov 16, 2026

Model weights

Pricing

Auto routing · per 1M tokens
Input
$0.30
Output
$1.00
Cache read
$0.03
Compare provider prices

Specifications

Context window
524.3K
Max output
131.1K
Parameters
594B / 17B
Total / active

Benchmarks

Sourced from LMArena.

Arena Score

1411.8

Overall Rank

#152 / 421

Votes

2,669

Confidence Interval

1400.0 - 1423.6

Category Scores

Coding

#107 / 416

619 votes

1484.8

Longer Query

#163 / 399

1,157 votes

1409.6

Creative Writing

#188 / 419

643 votes

1345.0

Instruction Following

#137 / 421

929 votes

1409.9

Hard Prompts

#138 / 421

1,665 votes

1438.3

Additional Categories
12

Expert

#66 / 371

251 votes

1487.9

Industry Business And Management And Financial Operations

#118 / 414

408 votes

1432.1

Industry Software And It Services

#123 / 421

960 votes

1463.9

English

#138 / 421

901 votes

1429.5

Hard Prompts English

#142 / 419

578 votes

1443.2

Multi Turn

#144 / 419

399 votes

1414.7

Non English

#150 / 421

1,768 votes

1397.2

Exclude Ties

#151 / 421

2,029 votes

1400.1

Industry Writing And Literature And Language

#157 / 420

823 votes

1382.9

Industry Life And Physical And Social Science

#161 / 419

380 votes

1422.7

Russian

#171 / 385

448 votes

1387.0

Industry Entertainment And Sports And Media

#173 / 419

709 votes

1360.2

Published 2026-11-11 · Matched as Inkling 2 Small · CC BY 4.0

LMArena Dataset

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Auto
Auto (ZDR)
$0.30$1.00$0.03N/AN/A
Auto
Auto (Anonymized)
$0.30$1.00$0.03N/AN/A
Tinfoil
Tinfoil
TEE verified
US
$0.315$1.05$0.0315N/AN/A
Privacy:ZDR, prompts not retainedNo training on prompts

Related text models

Compare Inkling 2 Small TEE with similar models from the same provider or model family.

Inkling Small

thinkingmachines/Inkling-Small

The direct-answer version of Thinking Machines' 276B-parameter open-weights multimodal Mixture-of-Experts model with 12B active parameters. It accepts text and images, and is designed for coding, tool use, instruction following, and general conversational work.

Inkling Small Thinking

thinkingmachines/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.

Inkling

thinkingmachines/inkling

The non-thinking version of Thinking Machines' 975B-parameter open-weights Mixture-of-Experts generalist with 41B active parameters. It gives faster direct answers across text, images, and audio, and is built for agentic coding, tool use, detailed instruction following, and long-context work.

Inkling Thinking

thinkingmachines/inkling:thinking

The thinking version of Thinking Machines' 975B-parameter open-weights Mixture-of-Experts generalist with 41B active parameters. It reasons natively over text, images, and audio, and is built for agentic coding, tool use, detailed instruction following, long-context work, and controllable thinking effort.

Kimi K3.1 Pro TEE

TEE/kimi-k3.1-pro

Kimi K3.1 Pro is Moonshot AI's multimodal model post-trained from Kimi K3 for long-horizon coding, agentic workflows, and knowledge work. It always reasons, and supports image input, tool calling, and structured output with a 262,144-token context window shared by input and output. This TEE (Trusted Execution Environment) deployment is verified through the selected provider: RedPill (Phala) attestation with signed completion receipts, or the official Tinfoil SDK's ATC/EHBP verification.

DeepSeek V4.1 Flash 1017 TEE

TEE/deepseek-v4.1-flash

DeepSeek V4.1 Flash's October 17, 2026 checkpoint (1017) supports text and image input, reasoning, tool calling, and structured output with a 1M-token context window. This route runs through Tinfoil attested inference inside a Trusted Execution Environment. Tinfoil lists this model as experimental, with availability subject to change.