AI BENCHY
Advertise here
#119

Inkling

Thinkingmachines Release: 2026-07-18 Tested on: 2026-07-18 15:45 thinkingmachines/inkling::low
975B total (41B active)MoEOpen source
(high) (medium) (low) (none)

Summary

Inkling scores 6.1 on AI BENCHY and ranks #119. It has 9.9 reliability, a 54.6% pass rate, $0.187 total cost, and 5.15s average response time.

What makes Inkling unique: It stands out most in General Intelligence, where it ranks #1, while Combined is its weakest area at #18.

Model facts

Researched on 2026-08-12

Reported
Parameters
975B total (41B active)
Architecture
MoE
Availability
Open source
License
Apache-2.0
Parameter source Availability source

Score

6.1

Consistency

8.6

Reliability

9.9

Total Cost (Current Price)

$0.187

Total Output Tokens

18,922

Total Input Tokens

109,884

Input Price

$1.000 / 1M

Output Price

$4.050 / 1M

Tests Correct

Wrong Tests: 12

Attempt pass rate: 54.6%

Flaky tests

4

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

5.15s

Response Time (max): 41.58s

Response Time (total): 113.39s

Wrong answer: 8 Did not follow instructions: 2 Invalid tool call: 2

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#119 Thinking Machines: Inkling

low
Cost
$0.002
Time
6.1s
Tokens
502 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-08-14 03:51 Re-test 6.1 10.0 $0.181 ↓ Compare
2026-07-18 15:45 Initial run 6.1 9.9 $0.187 Current run

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price
2026-07-18 15:56 $1.000 / 1M $4.050 / 1M

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

InklinglowvsQwen3.5-FlashmediumInklinglowvsTrinity Large ThinkingmediumInklinglowvsGemini 3.1 Flash LitenoneInklinglowvsQwen3.6 FlashnoneInklinglowvsQwen3.7 FlashnoneInklinglowvsGemma 4 31BnoneFree AvailableInklinglowvsgpt-oss-120bmediumInklinglowvsQwen3.5 Plus 2026-04-20noneInklinglowvsQwen3.5-35B-A3BmediumInklinglowvsNemotron 3 UltranoneFree Available

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 8.3 10.0
Coding 5.1 7.2
Combined 2.9 5.8
Data parsing and extraction 10.0 10.0
Domain specific 5.3 10.0
General Intelligence 10.0 10.0
Instructions following 9.8 10.0
Puzzle Solving 6.4 5.0
Tool Calling 3.0 10.0
Trivia 3.0 10.0

Compared models