GPT-6.1 Terra
- Provider
- OpenAI
- Status
- Current
- Context
- 1,050,000 tok
- Price
- $1 / $5 /MTok
GPT-6.1 Terra is OpenAI’s low-cost model for computer use, coding and office work, released on 26 October 2026 in the API and in ChatGPT Work and Codex (OpenAI). It costs $1/$5 per million tokens, half GPT-6.1 Sol’s price, with a 1,050,000-token context.
Artificial Analysis scored it 48 on the Intelligence Index v4.3.2 at max effort, at $0.40 per completed index task. OpenAI published no SWE-bench figure and no knowledge cutoff.
Quick specs
| GPT-6.1 Terra | |
|---|---|
| Released | 26 October 2026 |
| Input price | $1.00 / MTok |
| Cached input | $0.05 / MTok |
| Output price | $5.00 / MTok |
| Context window | 1,050,000 tokens |
| Max output | 128,000 tokens |
| Knowledge cutoff | Not published |
| AA Intelligence Index | 48 (max effort) |
| Cost per AA index task | $0.40 |
| Output speed | 90.9 tokens a second (Artificial Analysis) |
| API model ID | gpt-6.1-terra |
The context window and output limit are from the Amazon Bedrock model card. Token prices are text rates; tools such as web search and computer use are billed separately.
What OpenAI says is new
Built for computer use. OpenAI positions Terra for operating apps the way a person does, clicking and typing through interfaces, alongside coding and office tasks.
Speed. OpenAI says it finishes tasks more than three times faster than both GPT-6 Astra and GPT-6.1 Sol.
Cost per task. Against Claude Opus 5.5 at max effort, OpenAI claims more than three times the token efficiency at a quarter of the token price, for roughly nine times lower cost per task, as reported by The Decoder.
Safety. OpenAI’s system card rates it at the “Critical” cybersecurity threshold of its Preparedness Framework, with GPT-6 Astra’s safeguards, and puts its defence against indirect prompt injection at 98.8%, up from 96.2% for GPT-5.6 Terra, per The Decoder.
Benchmark performance
Independent, from Artificial Analysis:
| Measure | GPT-6.1 Terra | GPT-6 Sol | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|---|---|
| Intelligence Index v4.3.2 (max) | 48 | 48 | 52 | 53 |
| Cost per index task | $0.40 | $1.06 | $0.72 | $3.26 |
| Price (in/out per MTok) | $1 / $5 | $2 / $10 | $2 / $10 | $10 / $50 |
Artificial Analysis ranks it 25th of 221 models on its page. It used 130 million output tokens to run the index and scores six points above its predecessor, GPT-5.6 Terra, at $0.40 per task against $1.40.
Published by OpenAI:
| Benchmark | GPT-6.1 Terra | Comparison |
|---|---|---|
| Terminal-Bench 4.0 (max effort) | 50.3% | GPT-5.6 Terra 21.5% |
| Agents’ Last Exam | Level with GPT-6 Astra | At a tenth of Astra’s token price |
| Terminal-Bench-Science 0.1 | Within five points of Claude Opus 5.5 | At under a fifth of the estimated cost |
| OSWorld 2.1 (medium effort) | Above GPT-5.6 Terra’s best medium-effort result | No figure published |
OpenAI still recommends GPT-6.1 Sol and GPT-6 Astra for the most complex coding tasks.
How it compares
vs GPT-6 Sol
GPT-6 Sol scores the same 48 on the Artificial Analysis index at twice the token price ($2/$10), and $1.06 per completed task against Terra’s $0.40.
vs GPT-6.1 Sol
GPT-6.1 Sol scores four points higher (52) at $2/$10 and $0.72 per completed task. OpenAI says Terra is more than three times faster.
vs Claude Sonnet 5.5
Claude Sonnet 5.5 scores 56 on the same index at $2/$10, eight points higher, at $7.60 per completed task.
Known limitations
No SWE-bench figure. OpenAI published no SWE-bench Pro or Verified result.
Not in ChatGPT chat. At launch it is in ChatGPT Work and Codex, not regular ChatGPT conversations.
Mid-pack on the index. At 48, it sits eight to ten points below the top models on the Artificial Analysis index, and OpenAI points complex coding work to its larger models.
Specs withheld. Parameter count, architecture and knowledge cutoff are not published.
Frequently asked questions
When was GPT-6.1 Terra released?
26 October 2026, on the OpenAI API and in ChatGPT Work and Codex for Plus, Pro, Enterprise and Edu users.
How much does GPT-6.1 Terra cost?
$1 per million input tokens, $0.05 per million cached input tokens and $5 per million output tokens on the OpenAI API. Artificial Analysis measured $0.40 per completed Intelligence Index task.
Is GPT-6.1 Terra better than GPT-6.1 Sol?
Not on the independent index: Artificial Analysis scores GPT-6.1 Sol 52 and GPT-6.1 Terra 48. Terra costs half as much per token ($1/$5 against $2/$10), and OpenAI says it finishes tasks more than three times faster.
Can I use GPT-6.1 Terra in ChatGPT?
Yes, in ChatGPT Work and Codex on Plus, Pro, Enterprise and Edu plans. It was not available in regular ChatGPT chat at launch.
What is GPT-6.1 Terra for?
OpenAI built it for computer use, operating apps by clicking and typing, plus coding and office work, at a lower price than its Sol and Astra models.