AI BENCHY
Advertise here
#29

GPT-5.6 Terra

OpenAI Release: 2026-07-09 Tested on: 2026-07-09 21:31 openai/gpt-5.6-terra::high
~1T total (~60B active)MoEClosedEstimated
(high) (medium) (low) (none)

Summary

GPT-5.6 Terra scores 8.2 on AI BENCHY and ranks #29. It has 10.0 reliability, a 68.3% pass rate, $0.852 total cost, and 10.89s average response time.

What makes GPT-5.6 Terra unique: It stands out most in Domain specific, where it ranks #4, while Puzzle Solving is its weakest area at #17.

Model facts

Researched on 2026-08-12

Estimated
Parameters
~1T total (~60B 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

8.2

Consistency

9.6

Reliability

10.0

Total Cost (Current Price)

$0.852

Total Output Tokens

51,097

Total Input Tokens

34,218

Input Price

$2.500 / 1M

Output Price

$15.000 / 1M

Tests Correct

Wrong Tests: 7

Attempt pass rate: 68.3%

Flaky tests

1

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

Response Time (avg)

10.89s

Response Time (max): 91.49s

Response Time (total): 228.65s

Wrong answer: 7

Hamster playing table tennis

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

#29 GPT-5.6 Terra

high
Cost
$0.055
Time
30.2s
Tokens
3,748 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-08-14 02:25 Re-test 8.0 10.0 $0.418 ↓ Compare
2026-07-16 22:40 New test added 8.0 10.0 $1.055 Compare
2026-07-09 21:31 Initial run 8.2 10.0 $0.852 Current run

Run comparison

RunBenchmark coverageScoreConsistencyReliabilityTests CorrectFlaky testsTotal Output TokensTotal Input TokensTotal CostResponse Time (avg)
2026-07-09 21:31 · Initial run63/63 attempts8.29.610.014/21151,09734,218$0.85210.89s
2026-07-16 22:40 · New test added66/66 attempts8.09.310.014/22256,79181,047$1.05511.32s
Difference—+0.2+0.30.00-1-5694-46829-$0.203-436ms

Benchmark coverage differs: 63/63 attempts (Target: 3 repeats per test) versus 66/66 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.

These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price
2026-07-09 21:33 $2.500 / 1M $15.000 / 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 TerrahighvsInklingmediumGPT-5.6 TerrahighvsKimi K3maxGPT-5.6 TerrahighvsGLM 5.2highGPT-5.6 TerrahighvsGPT-5.2 ChatnoneGPT-5.6 TerrahighvsDeepSeek V4 Flash 0731highGPT-5.6 TerrahighvsStep 3.7 FlashmediumGPT-5.6 TerrahighvsDeepSeek V4 Flash 0731lowGPT-5.6 TerrahighvsQwen3.8 27BlowGPT-5.6 TerrahighvsMuse Spark 1.1highGPT-5.6 TerrahighvsQwen3.7 Plusmedium

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 8.3 10.0
Coding 7.6 7.2
Combined 10.0 10.0
Data parsing and extraction 10.0 10.0
Domain specific 5.3 10.0
General Intelligence 5.1 10.0
Instructions following 10.0 10.0
Puzzle Solving 7.7 10.0
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models