OpenAI launches GPT-6.1 Terra, a mid-tier model priced at $1 and $5 per million tokens
For the head of operations automation at a regional bank
OpenAI said in a post on its website that GPT-6.1 Terra, a mid-tier model, is available starting today to Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex, and to developers through the OpenAI API as gpt-6.1-terra; the post does not give a date, and The Register published its report on Oct. 26.21
OpenAI said in its announcement that Terra's standard API prices are 50% lower than those of GPT-5.6 Terra and GPT-6.1 Sol, and it named reading incoming invoices and entering their details into accounting software as an example of high-volume work developers can automate.2 The Register reported that Max Woolf, a data scientist, wrote in a blog post that inference providers will cut token prices to compete with each other, so OpenAI and Anthropic will have to lower theirs too; OpenAI did not give that as its reason.1
Standard API pricing is $1 per million input tokens, $0.05 per million cached input tokens and $5 per million output tokens, according to OpenAI's API documentation.3
The Register listed the other tiers as GPT-6 Luna at $0.10 and $0.50, GPT-6.1 Sol at $2 and $10, and GPT-6 Astra at $10 and $50 per million input and output tokens.1
Ted Sanders, a member of technical staff at OpenAI, said in a Hacker News comment that Terra is "probably better than Astra and Sol at technical communication, similar on vision and computer use (but much faster!), and a bit worse on most other things."1
On OpenAI's MMMU-Pro visual reasoning benchmark at max effort, InfoWorld reported Terra scored 85.5%, against 86.3% for Sol and 86.7% for Astra; the model accepts text and image input and produces text output.43
For banks and fintechs running document-heavy work such as invoice or loan-file extraction, vision quality is reportedly close to the pricier tiers at half Sol's token rate, so current contract pricing is worth renegotiating. OpenAI's figures are its own benchmarks, so teams should test accuracy and review effort on their own documents before switching.
Our take. OpenAI now sells a model that reads invoices at a tenth of Astra's price, and its own staff say it is a bit worse at most other things. Your contract renewal should cite the first fact. Your test plan should check the second.
Action. Operations automation: Run your invoice and loan-file extraction test set on GPT-6.1 Terra and compare accuracy and cost per document against your current model.
What our records say
OpenAI sells banks ChatGPT seats and API access, has added a financial-services edition and on-site engineering deals, and counts BBVA as a founding partner in its deployment company, with large deployments at BNY and Customers Bank.12 BNY has 20,000 employees building agents across 125-plus live use cases (as of Dec 12, 2025).3 75% of Customers Bank team members use OpenAI tools (as of Apr 27, 2026).4
From our records
- ChatGPT Enterprise (OpenAI)vendor · banks using it (14) Morgan Stanley, BNY, Customers Bank · alternatives Microsoft Copilot, Claude, Gemini Enterprise
- Claude (Anthropic)vendor · banks using it (21) Goldman, JPMorgan, BankUnited · alternatives Microsoft Copilot, ChatGPT Enterprise, Gemini Enterprise
Sources
- ReportOpenAI rolls out budget-priced mid-tier GPT-6.1 Terra Theregister
- Primary sourcehttps://openai.com/index/introducing-gpt-6-1-terra/ OpenAI
- Primary sourceGPT-6.1 Terra Model | OpenAI API OpenAI
- Also reported byOpenAI bets on AI economics with lower-cost GPT-6.1 Terra Infoworld