Back

OpenAI

GPT-6.1 Terra

ReasoningClosed weights
Intelligence
45.4
Rank 38 of 759 measured models
Output speed
90.9 t/s
Rank 229 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
$0.194
Rank 82 of 211 measured models

GPT-6.1 Terra (Xhigh) scores 45.4 on the Artificial Analysis Intelligence Index, ranking 38th of 759 measured models. The closest models on this index are Qwen3.8 Max (0902) (45.4), DeepSeek V4.1 Flash 1017 (Max) (45.1), and GLM-5.3 (44.8). At $2 per million tokens (3:1 input/output blended price), it is cheaper than Qwen3.8 Max (0902) ($3) and GLM-5.3 ($2.15), more expensive than DeepSeek V4.1 Flash 1017 (Max) ($0.525). With an output speed of 90.9 t/s, it is faster than Qwen3.8 Max (0902) (40.5 t/s), about as fast as GLM-5.3 (90.1 t/s), slower than DeepSeek V4.1 Flash 1017 (Max) (169 t/s).

Artificial Analysis Indices
Extra high
Intelligence45.4
Rank 38 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 & Accounting45.7
Rank 52 of 233 measured models
Strategy & Ops47.9
Rank 55 of 233 measured models
Legal47.9
Rank 52 of 233 measured models
Engineering46.4
Rank 39 of 232 measured models
Economics51.6
Rank 52 of 240 measured models
Standard Benchmarks
HLE43.7%
Rank 57 of 678 measured models
SciCode53.4%
Rank 79 of 245 measured models
AA-LCR81.7%
Rank 55 of 589 measured models
Terminal-Bench v446.5%
Rank 24 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