Back

OpenAI

GPT-6.1 Terra

ReasoningClosed weights
Intelligence
36.2
Rank 95 of 759 measured models
Output speed
91 t/s
Rank 226 of 418 measured models
Price/1M
$2
Rank 485 of 694 measured models
Context window
1M
Rank 92 of 1272 measured models
Cost per task
$0.0546
Rank 37 of 211 measured models

GPT-6.1 Terra (Low) scores 36.2 on the Artificial Analysis Intelligence Index, ranking 95th of 759 measured models. The closest models on this index are DeepSeek V4 Pro 0813 (36.0), Agnes 3.0 Flash (35.5), and Agnes 2.5 Pro Beta (35.2). At $2 per million tokens (3:1 input/output blended price), it is priced close to DeepSeek V4 Pro 0813 ($1.98), more expensive than Agnes 3.0 Flash ($0.075) and Agnes 2.5 Pro Beta ($0.15). With an output speed of 91 t/s, it is slower than DeepSeek V4 Pro 0813 (97.2 t/s) and Agnes 2.5 Pro Beta (108 t/s).

Artificial Analysis Indices
Low
Intelligence36.2
Rank 95 of 759 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 & Accounting38.2
Rank 94 of 233 measured models
Strategy & Ops41.9
Rank 89 of 233 measured models
Legal42.0
Rank 80 of 233 measured models
Engineering35.5
Rank 95 of 232 measured models
Economics44.4
Rank 93 of 240 measured models
Standard Benchmarks
HLE35.8%
Rank 122 of 678 measured models
SciCode52.4%
Rank 89 of 245 measured models
AA-LCR81.0%
Rank 65 of 589 measured models
Terminal-Bench v415.2%
Rank 77 of 232 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