whatmodel.sh / GPT-6.1 Terra

GPT-6.1 Terra

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

GPT-6.1 Terra is a closed-weights language model from OpenAI, released October 2026.

Model information sources

Release date ↗

83.8±0.9#3 of 731 LLMsfrom 10 benchmark results across 3 sources

Score within the LMArena (text) comparison set only.

$ how it compares

The LLMs ranked either side of GPT-6.1 Terra, out of 884 with a score, then more from OpenAI.

#modelscorewhereRAMreleased
1GPT-6 Astra OpenAI85±1api $20.00/M2026-09
2Claude Opus 5.5 Anthropic85±1api $8.00/M2026-09
3GPT-6.1 Terra OpenAI84±1api $2.00/M2026-10
4Gemini 4 Carbon Google83±1limited preview2026-10
5GPT-6.1 Sol OpenAI83±1api $4.00/M2026-09
more from OpenAI
11GPT-6 Sol OpenAI80±1api $4.00/M2026-09
12GPT-5.5 Pro OpenAI80±1api $67.50/M2026-04
13GPT-5.6 Sol OpenAI80±1api $8.00/M2026-07

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▄▅▅▄▄▃▁ #11 of 312 * · 11% of its score · 1 test

Trails Claude Fable 5, GPT-6 Sol, Claude Opus 5 here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Terminal-Bench 4.0 lab50.3% ✎▁·▁··▁▆▆▅██▃ beats 25/33very hard · avg 0.4%▇▇▇▇ 11%sharp · at its level · small run
coding▁▁▁▃▅▇█▇▅▄▃▁ #13 of 468 * · 14% of its score · 1 test

Trails Qwen3.8 Max (0902), Claude Mythos Preview (limited preview), GPT-6 Sol here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Arena WebDev (Code Arena)1672.7 ✎▁▁▃▂▅█▃▆▅▅▁▂ beats 117/129avg 982▇▇▇▇▇▇ 14%sharp
math▁▁▃▅█▇▄█▄▂▃▁ #4 of 429 · 24% of its score · 3 tests

Trails GPT-6.1 Sol here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
FrontierMath tier 4 v292.7% ✎▁▂▁▂█▅▂▃▃▁▁▁ beats 64/67very hard · avg 0.1%▇▇▇▇▇▇ 14%sharp
FrontierMath tiers 1–3 v290.9% ✎▂··▁▁▃▄▄█▆▄▃ beats 83/86very hard · avg 1.2%▇▇▇▇ 9.3%sharp · maxed out
OTIS mock AIME 2024–2025100.0% ✎▃▃▄▃▃▃▇▇█▃▂▅ beats 187/197very hard · avg 7.1%▇ 0.1%sharp · maxed out · noisy
other tests5 not in a task · 51% of its score
testresultvs the others in its rundifficultycountswhy
EBR-bench (Epoch)78.1% ✎▄▂·▆█▂█▆▄▆·▆ beats 25/25very hard · avg 0.3%▇▇▇▇▇▇▇▇ 20%sharp · at its level
Mystery Game Puzzles (Epoch)74.0% ✎█··▂▃▇▄█▂▄▃▁ beats 75/77very hard · avg 9.4%▇▇▇▇▇▇ 15%sharp · at its level · noisy
Furniture Assembly (Epoch)79.2% ✎█▆▆▆▆▂▆▂▂▂▃▅ beats 29/32very hard · avg 1.2%▇▇▇▇▇ 12%sharp · at its level
SimpleQA Verified57.3% ✎▁▁▃▂▃▃▆▆█▄▂▅ beats 64/81very hard · avg 8.0%▇ 2.6%at its level
GPQA Diamond94.3% ✎▁▁▁▁▂▄▆▄▅▇█▅ beats 203/211hard · avg 39%▇ 1.5%sharp · maxed out

$ where to run it

Through an API

api · $2.00/M · $1 in / $5 out per 1M tokens · openai.com ↗ · price sourceStandard USD per million uncached input/output tokens; cached input $0.05 excluded. Observed 2026-10-29.

$ how the score was worked out · Epoch AI — internal evaluatio… 75%, LMArena leaderboard dataset 14%, OpenAI vendor-reported benchm… 11%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Epoch AI — internal evaluations8CC BY 4.0
LMArena leaderboard dataset1CC BY 4.0
OpenAI vendor-reported benchmark scores for its own models (openai.com announcement posts and Deployment Safety Hub system cards) · lab-reported1Factual vendor-reported score (not licensed content)
prior · pull toward the middle
weighted average= 83.8

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json