‹ PublicAI Index
The LLM benchmark aggregator.
GPT-6.1 Terra
OpenAI
Strongest in Data analysis (#4 of 31), weakest in Legal (#149 of 173). Above par in 15 of 17 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and GPT-6 Astra and ahead of GPT-6 Sol and DeepSeek V4.1 Flash.
§ 1 · Profile
What it is good at
Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.
Core abilities61.4−3.9#8/1992/4
Data analysis61.4−3.8#4/311/1
General intelligence63.6−5.8#13/1782/3
Knowledge59.3−3.2#14/1971/3
Document parsing59.9−3.4#12/831/1
Coding58.8−7.4#16/1792/6
Agentic coding59.4−7.5#14/1732/5
Agents59−7.4#21/2983/6
Computer use62.7−5.4#7/681/2
Long-horizon work60.3−13.7#8/311/1
Web research50.5−13.4#22/351/2
Knowledge work62.1−8.6#35/2082/2
Reasoning55.7−9.7#36/1952/5
Reasoning57.9−11.1#32/1472/3
Professional47.8−14.6#127/1942/2
Document design60.3−8.4#7/311/1
Legal40.5−25.4#149/1731/1
§ 2 · Head to head
What it beats, and what beats it
The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for GPT-6.1 Terra, left for the other.
§ 3 · Sources
Where the numbers come from
7 publications, 21 figures. Every one links to the page it was read from.
Artificial Analysis Intelligence Index 51
GDPval-AA 1476
AA-Briefcase 1403
LiveBench 78.8LiveBench · Agentic 56.2LiveBench · Reasoning 88.7LiveBench · Coding 89.2LiveBench · Browser Use 93.1LiveBench · Data Analysis 82.3LiveBench · Research 79.1LiveBench · Design 79.7
SimpleBench 72.2%
Vals · Legal Research Bench 25.96%Vals · Harvey Legal Agent Benchmark 1.67%Vals · Vibe Code Bench 84% (OpenHands)
ParseBench · Overall 73.64%ParseBench · Tables 93.07%ParseBench · Charts 59.71%ParseBench · Content faithfulness 89.32%ParseBench · Semantic formatting 73.18%ParseBench · Visual grounding 52.94%
Badge
[](https://publicai.io/model-index/m/gpt-6-1-terra)