AI

GPT-6.1 Terra

Read full article at source ↗

Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 27/10/2026, 00:27:24 UTC.

Briefings

OpenAI released GPT-6.1 Terra, a model emphasizing visual reasoning and computer use, priced at $1/million input tokens and $5/million output. It is positioned similarly to Anthropic's Sonnet, offering a midrange option for technical communication and automated tasks, with a 10x price advantage over GPT-6 Luna. The release includes a detailed comparison grid of pricing and performance across reasoning levels.

Key points

• GPT-6.1 Terra combines detailed visual reasoning with precise computer use for tasks like drafting proposals and editing videos.
• The model costs $1/million input tokens and $5/million output, making it significantly cheaper than GPT-6.1 Sol while offering superior vision and computer capabilities.

Context

GPT-6.1 Terra represents a midrange option in OpenAI's LLM lineup, positioned similarly to Anthropic's Sonnet. It is specifically noted for excelling at communication and computer-based tasks, such as interpreting technical drawings and automated repetitive work, which benefits from less expensive models compared to top-tier competitors.

Limitations

The evidence is limited to a single article from October 2026. Pricing data is provided in cents per million tokens and dollars per million tokens, but specific usage limits or performance benchmarks beyond the provided grid are not detailed. The article does not contain direct quotes from OpenAI or external third-party evaluations, and the 'pelican' pricing data appears to be a metaphorical comparison rather than a literal cost metric.

Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 27/10/2026, 01:06:14 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.

View archived briefing on IPFS
bafybeifwlx7jtiab76ztnm73wotp4f4bc7bpdcjmqpbbmfi7q74br4hhim