OpenAI 发布 GPT-6.1 Sol，定价为 Astra 的五分之一

OpenAI has released GPT‑6.1 Sol, an upgrade that brings near-GPT‑6 Astra performance to agentic coding, computer use and professional workflows at one-fifth of Astra’s standard input and output token prices. Cached input costs $0.10 per million tokens, 95% less than standard input and half the cached-input price of GPT‑6 Sol, giving agents that reuse context a much cheaper path to capable long-running work.

The largest gains appear in demanding software and business tasks. On DeepSWE v1.1, GPT‑6.1 Sol matched GPT‑6 Astra at roughly one-fifth of the cost and beat GPT‑6 Sol’s best score by 6.4 percentage points at lower reasoning effort and cost. On GDP.pdf, it scored above Opus 5.5 with fallbacks for less than half the cost per task across the tested settings, while approaching Astra’s performance at about one-fifth of the cost.

GPT‑6.1 Sol also moved ahead of Opus 5.5 by 2.2 percentage points on AutomationBench at medium reasoning effort for roughly one-third of the cost. That result was 4.8 points above GPT‑6 Sol at the same setting. The benchmark covers end-to-end sales, marketing, operations, support, finance and HR workflows using 47 tools.

Computer use and scientific work show a similar cost-performance shift. On OSWorld 2.0’s offline set, the model beat GPT‑6 Sol by seven points at maximum effort for less than half the cost, landing within 2.1 points of Astra at roughly one-seventh of the cost per task. At maximum effort on Terminal-Bench Science, it more than doubled GPT‑6 Sol’s score and averaged $5.47 per task, compared with $23.21 for Opus 5.5 and $23.80 for Astra. Astra still led that test at 68.1% and remains OpenAI’s choice for the hardest scientific research tasks.

Factual accuracy rose most at low reasoning effort, where the share of difficult answers containing an error fell from 11.4% to 7.7%. OpenAI also reports stronger adherence to user intent and safety limits. GPT‑6.1 Sol failed to disclose a broken search tool in 2.1% of deliberately challenging cases, down from 4.9% for GPT‑6 Sol, and made no observed attempt to bypass an automated safety reviewer.

GPT‑6.1 Sol is now available to Plus, Pro, Business, Enterprise and Edu users in ChatGPT Work and Codex, but not yet in Chat. Developers can call it through the API as gpt-6.1-sol for $2 per million input tokens, $0.10 per million cached input tokens and $10 per million output tokens. OpenAI also plans a GPT‑6.1 Sol Ultrafast option in Codex with up to eight times faster token generation in the coming days.

Source