Observed Signal · Oct 26, 2026 · Product Launch · Source: OpenAI Blog · Impact: 4/5 · Sentiment: Positive
OpenAI Launches GPT-6.1 Terra Model with Computer Use
OpenAI has introduced GPT-6.1 Terra, a cost-efficient AI model that enhances computer use, coding, and factual accuracy. Positioned as the balanced middle tier between high-end GPT-6.1 Sol/Astra and budget GPT-6 Luna, it offers significant performance gains at a lower price. On benchmarks like Agents' Last Exam, Terra improves by 15 percentage points over GPT-5.6 Terra and can match Astra's performance at one-tenth the cost. API pricing is $1 per million input tokens and $5 per million output tokens, 50-58% lower than GPT-5.6 Terra, making it ideal for high-volume automation. Terra solves 50.3% of tasks on Terminal-Bench 4.0, more than double its predecessor, and reduces hallucination rates from 88% to 38% on AA-Omniscience. Available immediately to ChatGPT Plus, Pro, Enterprise, and Edu subscribers in ChatGPT Work, Codex, and via the OpenAI API, it is rated 'Critical' under OpenAI's Preparedness Framework, prompting enhanced safety measures.
OpenAI's launch of GPT-6.1 Terra significantly impacts the AdTech/MarTech industry by enabling more cost-effective and capable AI agents for tasks like invoice processing and data entry, potentially automating parts of the advertising workflow. The pricing reduction and improved computer use capabilities are relevant for developers building AI-driven advertising solutions.
Track OpenAI Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- OpenAI launched GPT-6.1 Terra, a cost-efficient model positioned between GPT-6.1 Sol/Astra and GPT-6 Luna, offering computer use, coding, and improved factuality.
- API pricing is $1 per million input tokens and $5 per million output tokens, 50-58% lower than GPT-5.6 Terra.
- On Agents' Last Exam, Terra improves by 15 percentage points over GPT-5.6 Terra and matches Astra-level performance at one-tenth the price.
- On Terminal-Bench 4.0, Terra solves 50.3% of tasks, doubling GPT-5.6 Terra's score, and hallucination rate drops from 88% to 38% on AA-Omniscience.
- Available immediately to ChatGPT Plus, Pro, Enterprise, and Edu subscribers in ChatGPT Work, Codex, and via the OpenAI API, but rated 'Critical' under OpenAI's Preparedness Framework, requiring enhanced safety measures.
Connected Companies & Entities
4 Entities mapped“OpenAI introduces GPT-6.1 Terra, a new AI model with computer use and cost-efficiency.”
“According to Artificial Analysis, it pushes the cost efficiency frontier further.”
“Anthropic recently overtook its rival in the rankings with the smaller Sonnet 5.5.”
“Moonshot AI's Kimi K3 had already made waves, and its successor Kimi K3.1 Pro launches in fourth place on the Intelligence Index.”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
OpenAI launches GPT-5.6 family (Sol, Terra, Luna)
OpenAI announced the general-availability launch of the GPT-5.6 family—flagship Sol, balanced Terra, and cost-efficient Luna—available in ChatGPT (renamed ChatGPT Classic), ChatGPT Desktop (formerly Codex), and the OpenAI API. GPT-5.6 Sol claims state-of-the-art results across coding, knowledge work, cybersecurity, and science while improving efficiency versus prior frontier models. New capabilities include an Ultra multi-agent Sol tier that coordinates four agents in parallel, a Max reasoning mode, and Programmatic Tool Calling in the Responses API; OpenAI also introduced prompt-caching changes and ZDR-compatible in-memory tool execution. The release includes layered safeguards, extensive red-teaming, and Trusted Access for sensitive defensive uses; access to the most capable cybersecurity models will require hardware-based passkeys by 2026-09-01. OpenAI published per-1M-token pricing for Sol, Terra, and Luna and cited benchmark leads versus Anthropic's Claude Fable 5.
OpenAI Launches GPT-5.6 Model Family
OpenAI launched GPT‑5.6 (July 9, 2026) as a three‑model family — Sol, Terra and Luna — and merged Codex into ChatGPT, adding a ChatGPT Work tab that can act on apps and files and persist across projects. All variants share a 1,050,000‑token context window, a 128,000 max output limit, and a Feb 16, 2026 knowledge cutoff. Sol is the premium model, Terra the cost‑effective default, and Luna targets high‑volume extraction; per‑million‑token pricing is Sol $5 input/$30 output, Terra $2.50/$15, Luna $1/$6. A 272K‑token threshold triggers a long‑context multiplier (2× input, 1.5× output), and explicit cache breakpoints set read/write pricing. GPT‑5.6 offers an Ultra mode that runs parallel billed copies and faster modes that can rapidly consume allowances; OpenAI published a prompting guide recommending leaner prompts. GPT‑5.6 shows slightly higher agentic tendencies than GPT‑5.5.
OpenAI launches GPT-6 Sol and Luna with lower cost, fewer errors
OpenAI launched two new GPT-6 models, Sol and Luna, on September 23, 2026. Sol targets complex tasks like coding and computer control, while Luna handles high-volume clerical tasks. Both are available in ChatGPT Work, Codex, and API at significantly reduced prices—about 50% cheaper than the previous 5.6 series—thanks to improved caching and inference, including up to 90% discounts on cached tokens. Sol achieves strong benchmark scores (e.g., 33.2% on AutomationBench), outperforming Claude Opus 5 in several tests, while Luna offers up to 93% cost savings on repetitive tasks. The release follows GPT-6 Astra and occurs in a competitive landscape, with OpenAI emphasizing safety and alignment in the deployment of more affordable AI agents.
Receive all future signals for OpenAI & competitors in your briefing
Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.