OpenAI introduces GPT-6.1 Terra with lower API pricing and real-world coding tests
OpenAI introduced GPT-6.1 Terra, about 50% cheaper than GPT-5.6 Terra, with benchmark results from fixing 105 bugs in real repositories.
OpenAI introduced GPT-6.1 Terra, and the new model's cost turned out to be significantly lower than expected.
API usage costs:
$1 per million input tokens. $5 per million output tokens. $0.05 per million cached input tokens.
As a result, GPT-6.1 Terra is approximately 50% cheaper than its predecessor, GPT-5.6 Terra.
The new model has already been tested on real software development tasks.
For testing, two repositories were used, into which 105 bugs were intentionally introduced. The models were tasked with finding and fixing as many of them as possible.
Test results:
- GPT-6.1 Terra (max) — 40.7
- GPT-5.6 Terra (max) — 32
- GPT-5.6 Terra (xhigh) — 20
- GPT-5.6 Terra (high) — 18
- GPT-6.1 Terra (xhigh) — 11.3
- GPT-6.1 Terra (high) — 9.7
GPT-6.1 Terra showed good efficiency both in cost and task execution time. However, when the reasoning level was reduced from max, the model's performance dropped sharply, falling significantly below the results of Haiku 5.5 (xhigh).
Individual run results:
- GPT-6.1 Terra (max) — 34, 42, 46.
- GPT-6.1 Terra (xhigh) — 11, 9, 14.
- GPT-6.1 Terra (high) — 11, 8, 10.
P. S. Yes, Sonnet 5.5 (max) still wins, performing seven times more iterations than Astra. The matter is not the model's outstanding abilities, but that it persistently continues working and does not stop.
Why it matters
AnalysisThe release suggests a possible cost-performance shift for developer-facing AI models, while reduced reasoning levels can sharply degrade coding reliability.
Discuss in community
Share your questions and insights with developers
Related in Ecosystem
Practical guides, core terminology, and agent skills
OpenAI releases GPT-6.1 Terra: near Sol performance, half the price, faster coding
GPT-6.1 Terra nearly matches GPT-6.1 Sol on benchmarks while being twice as cheap, with stronger visual understanding, computer use, more natural text, and faster coding.
OpenAI Releases GPT-6.1 Terra
OpenAI has released GPT-6.1 Terra, with the official announcement available on the company's website.
OpenAI GPT (Flagship Models of the GPT Series)
The primary universal line of large language models from OpenAI (GPT-4, GPT-4o). Optimized for complex text analysis, programming, creativity, and daily intellectual tasks.
Codex Skills for Beginners: Automating AI Agent Workflows
A comprehensive hands-on guide to building and executing Codex Skills: SKILL.md anatomy, explicit vs. auto-triggering, skills vs. scripts vs. tools, Skills API, and safety boundaries.
GPT Mini (Compact Models of the GPT-4o Mini Series)
An economical and ultra-fast version of OpenAI's flagship models. Optimized for mass daily tasks, instant responses, text classification, and limit conservation.
How to Integrate Monobank Acquiring Without a Developer via Vibe-Coding
A step-by-step guide to integrating Monobank online payments: FOP compliance, terminal setup, ECDSA webhook verification, robust UX, and security test matrix.