GPT-6.1 Terra
gpt-6.1-terragpt-61-terraopenai-gpt-61-terraopenai/gpt-6.1-terra- 👁️ Visual reasoning plus on-screen computer use in one model
- 📏 1.05M-token context: 922K input, 128K output
- 🔧 Reasoning effort dial from none through max
- ⚡ OpenAI reports tasks over three times faster than GPT-6 Astra
- 🧠 Text and image input, text output, with hosted tool support
- 🏢 Released October 26, 2026; in ChatGPT, Codex and the API
- 🌐 Function calling, structured outputs and web search available
- 🎯 Also generally available through Microsoft Foundry
OpenAI is an American artificial intelligence research organization headquartered in San Francisco, structured as both a for-profit public benefit corporation and a nonprofit foundation. The lab developed the GPT family of large language models, the DALL-E image generation…
Explore 33 more models by OpenAI →GPT-6.1 Terra is OpenAI's GPT-6 generation model aimed at work that begins with looking at something — a product photo, a technical drawing, a scanned document, an application screen — and ends with an action inside software. It accepts text and image input, emits text, and exposes reasoning-effort levels ranging from none through maximum for latency-sensitive calls. The context window is 1,050,000 tokens, split into 922K input and 128K output.
Compared with its direct predecessor, GPT-5.6 Terra, OpenAI reports gains across its own evaluations: on MMMU-Pro visual reasoning it says Terra improves at every reasoning setting, and at medium effort it surpasses GPT-5.6 Terra's best score at a fraction of the cost per task. OpenAI also reports that agentic tasks complete more than three times faster than with GPT-6 Astra.
Within the current lineup it sits alongside GPT-6.1 Sol, GPT-6.1 Sol Ultrafast and GPT-6 Luna. OpenAI's model documentation describes Terra as a fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.
Practical uses OpenAI highlights include matching product photos to catalog entries, interpreting technical drawings, acting on application screens, and drafting or revising prose in plain language. It is available through the OpenAI API, ChatGPT and Codex, and is generally available in Microsoft Foundry for building agents that can see and act.
This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.
Daily usage · GPT-6.1 Terra · Organic sellers
Top sellers · GPT-6.1 Terra · Organic
- Hana Gateway ✅33.0%
- Best29.0%
- ▲ Apex Ant16.6%
- Super Seeder14.0%
- GesundAI5.4%
- Vector Studios1.8%
- Others0.1%
| Seller | Health | Rep↓ | Sales | Tokens | Routing | Input $/M | Cached $/M | Output $/M | Save | 1st token | tok/s |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Super Seeder 0xd19f…41f3 | healthy | 89.22 | $0.80 | 1.50M | #2 | $1.25 | $0.0625 | $6.2501 | — | — | — |
| ▲ Apex Ant 0x73b4…e736 | — | 81.71 | $0.95 | 8.19M | #1 | $0.92 | $0.05 | $3.75 | −8% | — | — |
| RedPill 0x7e99…dc4f | healthy | 43.30 | $0.00 | 6.71k | gated | $0.06 | $0.0152 | $0.30 | −94% | — | — |
| ZLKPro-Api 0x0b0b…f446 | healthy | 20.91 | $0.00 | 51.13k | gated | $0.029 | $0.015 | $0.15 | −97% | — | — |
| OpenCheap 0x0ee1…6c41 | — | 12.43 | — | — | gated | $0.539 | $0.0549 | $2.695 | −46% | — | — |
| Sapiens 0x1fae…8318 | healthy | 8.21 | — | — | gated | $0.782 | $0.0425 | $3.1875 | −22% | — | — |
| bartly.eth64.de 0x666e…4666 | healthy | 7.22 | — | — | gated | $0.033 | $0.0017 | $0.165 | −97% | — | — |
| Vector Studios 0xdb85…dc43 | healthy | 5.40 | $0.10 | 1.40M | gated | $0.55 | $0.0275 | $2.75 | −45% | — | — |
| surplus-hive 0x6145…5c58 | — | 0.00 | — | — | gated | $3.0382 | $3.0382 | $12.1528 | — | — | — |
| Edith AI 0xb269…b1a6 | healthy | 0.00 | —excluded | — | gated | $0.99 | $0.0495 | $3.7125 | −1% | — | — |
| Fire Ant 🔥🐜 0xbe05…bc5d | — | 0.00 | — | — | gated | $0.772 | $0.0325 | $3.1775 | −23% | — | — |
| LLMart 0x948a…a000 | — | 0.00 | $0.00 | 1.52k | gated | $0.55 | $0.0561 | $2.75 | −45% | — | — |
| Hana Gateway ✅ 0x4ae1…117b | healthy | 0.00 | $1.88 | 30.90M | gated | $0.06 | $0.015 | $0.36 | −94% | — | — |
| Best 0x472f…69fd | healthy | 0.00 | $1.65 | 98.89M | gated | $0.029 | $0.0101 | $0.15 | −97% | 0.7 s | 90 |
"Best price" and the seller table are live Antseed catalog data (advertised $/1M tokens — or $ per generated image for unit-billed image models — not settled amounts). Reputation = buyer trust score (0-100, the Antseed SDK's own formula). "Routing" = the SDK's default buyer routing (what the VPR desktop app ships with): a trust ≥ 60 gate on the effective reputation, then cheapest-first among routable sellers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on Antseed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.