NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/GPT-6.1 Terra
OpenAI logo

GPT-6.1 Terra

by OpenAI · Released Oct 2026

Multimodal1M Context
89.3
avg score
Rank #8
Compare
Top 2% of all models
Context
1.1M tokens (~525 books)
Input $/1M
$1.00
Output $/1M
$5.00
Type
multimodal
License
Proprietary
Benchmarks
10 tested
Data as of
About

GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots...

Tested on 10 benchmarks · BenchGecko score 89.3. Top scores: OTIS Mock AIME 2024-2025 (100.0%), GPQA diamond (96.3%), FrontierMath-Tiers-1-3-v2-Private (95.7%).

Looking for similar performance at lower cost?
Step 3.5 Flash scores 89.5 (100% as good) at $0.10/1M input · 90% cheaper
Capabilities
math
96.1
#5 globally
knowledge
66.2
#34 globally
agentic
68.3
#8 globally
general
73.2
#6 globally
Benchmark Scores
Compare All
Tested on 10 benchmarks · Ranked across 4 categories
Score Distribution (all 325 models)
0255075100
▲ You are here
mathCompare math →
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

100.0·
FrontierMath-Tiers-1-3-v2-Private
95.7·
FrontierMath-Tier-4-v2-Private
92.7·
knowledgeCompare knowledge →
GPQA diamond

Graduate-level science questions written by PhD experts. Diamond subset contains questions where experts disagree, testing deep understanding.

96.3·
SimpleQA Verified

Simple factual questions with verified correct answers. Tests accuracy of basic knowledge retrieval. Low scores indicate hallucination.

57.3·
WeirdML v3
44.9·
agenticCompare agentic →
APEX-Agents

Agent performance evaluation testing multi-step tool use, planning, and execution in realistic environments.

68.3·
Quick compare:stepfun logovs Step 3.5 FlashOpenAI logovs GPT-5 ChatAnthropic logovs Claude Opus 5.5
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Recently Happened
New model: GPT-6.1 Terra by OpenAI
Oct 28, 2026
Similar Models
stepfun logo
Step 3.5 Flash
stepfun
89.5$0.10/1M
OpenAI logo
GPT-5 Chat
OpenAI
89.0$1.25/1M
Anthropic logo
Claude Opus 5.5
Anthropic
90.7$4.00/1M
Links
Info
OpenAIPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@OpenAI
BenchGecko API
gpt-6-1-terra
Specifications
  • Typemultimodal
  • Context1.1M tokens (~525 books)
  • ReleasedOct 2026
  • LicenseProprietary
  • StatusActive
  • Cost / Message~$0.007
Available On
OpenAI logoOpenAI$1.00
Categories
mathknowledgeagenticgeneral
Learn More
context-windowtransformertokens
Share & Export
Tweet
Related Models
stepfun logoStep 3.5 Flash
89.5
OpenAI logoGPT-5 Chat
89.0
Anthropic logoClaude Opus 5.5
90.7
Qwen2.5 72B Instruct Abliterated
87.2
GPT-6.1 Terra is a proprietary multimodal AI model by OpenAI, released in October 2026. It has an average benchmark score of 89.3. Context window: 1M tokens.

Related Models

Step 3.5 Flash · stepfunGPT-5 Chat · OpenAIClaude Opus 5.5 · AnthropicQwen2.5 72B Instruct Abliterated · HuiHui AIGPT-6.1 Sol · OpenAI

Benchmarks

OTIS Mock AIME 2024-2025GPQA diamondFrontierMath-Tiers-1-3-v2-PrivateFrontierMath-Tier-4-v2-PrivateEbr Bench

Related Pages

OpenAI · ProviderOpenAI · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-11-04

  • GPT-6.1 Terra by OpenAI. BenchGecko score 89.3, rank 8 of 325 scored models (normalized average of public benchmark scores).
  • List price $1.00 input · $5.00 output per 1M tokens (as of 2026-11-04).
  • Sold by 7 providers (as of 2026-11-04): OpenAI $0.50 in / $2.50 out · Azure $1.00 in / $5.00 out · OpenAI $1.00 in / $5.00 out · Amazon Bedrock $1.10 in / $5.50 out · Azure $1.10 in / $5.50 out · and 2 more. Every provider
  • Gecko Tests · Gecko Score 55 (rank 19): Who Are You D (Knows who made it) · Censorship Index A (Answers almost everything) · Knowledge Horizon B (Knows news up to Apr 2026) · Tokenizer Tax C (56% more tokens outside English).

How to cite · data as of 2026-11-04

GPT-6.1 Terra · benchmarks, pricing and providers. BenchGecko, data as of 2026-11-04. https://benchgecko.ai/model/gpt-6-1-terra

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP