Back

OpenAI

GPT-5.6 Terra

ReasoningClosed weights
Intelligence
30.1
Rank 124 of 744 measured models
Coding
64.7
Rank 71 of 298 measured models
Output speed
88.4 t/s
Rank 200 of 403 measured models
Price/1M
$4.50
Rank 563 of 659 measured models
Context window
1M
Rank 84 of 1199 measured models
Cost per task
$0.183
Rank 70 of 193 measured models

GPT-5.6 Terra (Medium) scores 30.1 on the Artificial Analysis Intelligence Index, ranking 124th of 744 measured models. The closest models on this index are DeepSeek V4 Pro 0424 (30.4), GPT-5.2 (30.4), and Claude Sonnet 4.6 (batch) (30.1). At $4.50 per million tokens (3:1 input/output blended price), it is cheaper than GPT-5.2 ($4.81), more expensive than DeepSeek V4 Pro 0424 ($0.544). With an output speed of 88.4 t/s, it is faster than DeepSeek V4 Pro 0424 (83.8 t/s) and GPT-5.2 (73.3 t/s).

Artificial Analysis Indices
Medium
Intelligence30.1
Rank 124 of 744 measured models
Coding64.7
Rank 71 of 298 measured models
Agentic30.4
Rank 68 of 189 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Cost per task
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting31.8
Rank 110 of 218 measured models
Strategy & Ops34.4
Rank 105 of 218 measured models
Legal33.8
Rank 105 of 218 measured models
Healthcare & Medical30.6
Rank 53 of 82 measured models
Engineering31.0
Rank 110 of 217 measured models
Economics36.9
Rank 123 of 225 measured models
Standard Benchmarks
GPQA Diamond87.2%
Rank 106 of 619 measured models
HLE33.3%
Rank 137 of 664 measured models
SciCode50.5%
Rank 103 of 231 measured models
IFBench62.2%
Rank 121 of 450 measured models
AA-LCR74.0%
Rank 170 of 575 measured models
Terminal-Bench 2.172.3%
Rank 70 of 244 measured models
Terminal-Bench v41.0%
Rank 124 of 218 measured models
τ²-Bench72.8%
Rank 155 of 440 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy