AI BENCHY
Advertise here
#151

GPT-5.6 Luna

OpenAI Release: 2026-07-09 Tested on: 2026-07-16 22:36 openai/gpt-5.6-luna::none
~400B total (~17B active)MoEClosedEstimated
(high) (medium) (low) (none)

Summary

GPT-5.6 Luna scores 5.4 on AI BENCHY and ranks #151. It has 10.0 reliability, a 34.9% pass rate, $0.142 total cost, and 1.50s average response time.

What makes GPT-5.6 Luna unique: It is notably fast compared with similar models.

Model facts

Researched on 2026-08-12

Estimated
Parameters
~400B total (~17B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. Vendor does not disclose the architecture or count.

Parameter source Availability source

Score

5.4

Consistency

8.8

Reliability

10.0

Total Cost (Current Price)

$0.142

Total Output Tokens

6,709

Total Input Tokens

101,323

Input Price

$1.000 / 1M

Output Price

$6.000 / 1M

Tests Correct

Wrong Tests: 16

Attempt pass rate: 34.9%

Flaky tests

3

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

Response Time (avg)

1.50s

Response Time (max): 10.57s

Response Time (total): 32.91s

Wrong answer: 14 Extra formatting: 1 Invalid tool call: 1

Hamster playing table tennis

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

#151 GPT-5.6 Luna

none
Cost
$0.016
Time
15.8s
Tokens
2,685 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-08-14 02:19 Re-test 5.4 10.0 $0.015 ↓ Compare
2026-07-16 22:36 New test added 5.4 10.0 $0.142 Current run
2026-07-09 21:15 Initial run 5.3 10.0 $0.047 Compare

This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price
2026-07-09 21:18 $1.000 / 1M $6.000 / 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

GPT-5.6 LunanonevsKAT-Coder-Air V2.5lowGPT-5.6 LunanonevsMiMo-V2.5-PrononeGPT-5.6 LunanonevsLaguna S 2.1mediumFree AvailableGPT-5.6 LunanonevsKimi K2.5noneGPT-5.6 LunanonevsDeepSeek V4 Flash 0423noneGPT-5.6 LunanonevsDeepSeek V4 Flash 0731noneGPT-5.6 LunanonevsQwen3.8 27BnoneGPT-5.6 LunanonevsLaguna S 2.1highFree AvailableGPT-5.6 LunanonevsQwen3.6 27BnoneGPT-5.6 LunanonevsInkling Smallnone

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 4.8 10.0
Coding 3.8 7.2
Combined 3.2 9.1
Data parsing and extraction 10.0 10.0
Domain specific 2.9 7.2
General Intelligence 5.0 10.0
Instructions following 7.1 5.8
Puzzle Solving 5.3 10.0
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models