GPT-6.1 Terra
GPT-6.1 Terra is a chat model from OpenAI, tracked here with measured throughput of roughly 91 output tokens per second and a time to first token near 1.8 seconds.
API Pricing
| Provider | Input / 1M | Output / 1M | Cached / 1M |
|---|---|---|---|
| $0.500 | $2.50 | $0.025 | |
| $1.00 | $5.00 | $0.050 |
Prices updated daily. Last check: Oct 27, 2026
Compare API pricing for every OpenAI model →GPT-6.1 Terra pricing by provider
Input, output, and batch rates, plus alternatives, for one provider at a time.
Performance & Benchmarks
Source: Artificial Analysis →Reasoning & Knowledge
- Humanity's Last Exam46.2%
Coding
- SciCode55.1%
Instruction & Long Context
- Long-Context Reasoning82.7%
Benchmarks measured Oct 2026. Scores are independent evaluations, not vendor-reported.
Model Details
General
- Creator
- OpenAI
- Modalities
- Text
Capabilities
Strengths & Limitations
Strengths
- Measured output throughput of approximately 90.73 tokens per second in Artificial Analysis testing
- Time to first token of roughly 1,829 ms, suitable for streaming chat interfaces
- Published by OpenAI, whose chat models are widely supported by existing SDKs, agent frameworks, and tooling
- Standard chat-completion interface, so integration patterns carry over from other GPT-series endpoints
- Independent third-party benchmark data available rather than vendor-only performance claims
- Listed with live, provider-by-provider pricing in the table on this page for direct cost comparison
Limitations
- We do not currently track a verified context window size for this model, so long-document limits should be confirmed with the provider
- Input and output modalities beyond text chat are not confirmed in our data
- Time to first token near 1.8 seconds is noticeable in latency-sensitive, turn-by-turn interfaces without streaming
- Throughput and latency figures come from a single benchmark source and will vary by provider, region, and load
- Quality benchmarks (reasoning, coding, multilingual) are not available in our dataset for this entry
Key Features
About GPT-6.1 Terra
Common Use Cases
GPT-6.1 Terra is positioned as a general chat model, which makes it a candidate for assistant interfaces, drafting and rewriting workflows, summarization, customer-facing Q&A over supplied context, and as the reasoning step inside retrieval-augmented or agent pipelines that speak the standard chat-completion protocol. Its measured ~91 tokens per second output rate is comfortable for streamed responses where the user reads along as text appears, and for background or batch jobs where total wall-clock time matters more than first-token responsiveness. The ~1.8 second time to first token is worth accounting for in voice or rapid-turn interfaces, where perceived delay before the first word is the dominant factor; such designs usually compensate with streaming, prefetching, or a smaller router model for short turns. Because our entry does not carry a confirmed context window or modality set, teams planning long-document ingestion or image input should verify those specifics with the serving provider before committing.
Frequently Asked Questions
How much does GPT-6.1 Terra cost?
Pricing varies by provider and by pricing type — input versus output tokens, cached input, and any batch or committed-use discounts are each priced differently, and rates change over time. Check the pricing table on this page for current per-provider rates rather than relying on a fixed figure.
What is GPT-6.1 Terra best used for?
It is a chat model, so it fits assistant and conversational interfaces, summarization and drafting, Q&A over supplied context, and as the generation step in retrieval-augmented or agentic pipelines built on the chat-completion API. Its roughly 91 tokens per second output rate works well for streamed responses and for batch jobs where total completion time is the main concern.
How fast is GPT-6.1 Terra?
Artificial Analysis measured approximately 90.73 output tokens per second with a time to first token of about 1,829 milliseconds. Both numbers depend on the serving provider, region, prompt size, and concurrent load, so your observed figures may differ.
What context window does GPT-6.1 Terra support?
We do not currently track a verified context window value for this entry. That means the figure is unconfirmed on our side, not that the model is limited — confirm the supported context length with the provider you plan to use.
Can GPT-6.1 Terra accept images or other non-text input?
Our metadata does not confirm modalities beyond text chat for this model. If image, audio, or document input matters to your workload, verify support directly with the provider listed in the pricing table before building against it.
How does GPT-6.1 Terra compare to other OpenAI chat models?
The comparison that our data supports is on serving performance and price: roughly 91 output tokens per second and a ~1.8 second first-token delay, weighed against whatever rate appears in the pricing table on this page. We do not hold quality benchmark scores for this entry, so capability comparisons against sibling models should be based on your own evaluation on representative prompts.