- Published on
OpenAI Releases GPT-6.1 Sol: Near-Astra Coding and Computer Use at One-Fifth of Astra’s Token Price
- Authors
- Name
- DeepHorus
OpenAI has released GPT-6.1 Sol, a refreshed version of GPT-6 Sol, the mid-tier entry in the GPT-6 model family. The company's claim is specific: near-Astra results on agentic coding, computer use, and professional work, delivered at one-fifth of GPT-6 Astra's standard input and output rates. Cached input drops to $0.10 per million tokens, a 50% cut versus GPT-6 Sol. The model is deployable today as a hosted endpoint in the OpenAI API under the name gpt-6.1-sol, as well as in ChatGPT Work and Codex.
The release lands within a three-tier GPT-6 lineup. GPT-6 Astra is priced at $10 input, $50 output, and $1 cached input per million tokens; GPT-6.1 Sol at $2, $10, and $0.10; and GPT-6 Luna at $0.10, $0.50, and $0.01. The cached price matters most for agents, which resend the same system prompt, tool schemas, and history at every step, and cached reads now cost 5% of the uncached input rate, down from 10% on GPT-6 Sol.
On benchmarks, all figures vendor-reported in OpenAI's launch post, Sol matches GPT-6 Astra on DeepSWE v1.1 coding at roughly one-fifth the cost and beats GPT-6 Sol's best score by 6.4 percentage points at a lower reasoning effort. On GDP.pdf, covering complex professional PDFs, it outperforms Claude Opus 5.5 with fallbacks at less than half the cost per task. On AutomationBench 1.0.6 it scores 2.2 points above Opus 5.5 at medium effort, at about one-third the cost.
For computer use, Sol gains 7 points over GPT-6 Sol on the OSWorld 2.0 offline set at maximum effort, landing within 2.1 points of Astra at roughly one-seventh the cost per task. On Terminal-Bench Science 0.1, it more than doubles GPT-6 Sol's score, averaging $5.47 per task versus $23.21 for Opus 5.5 and $23.80 for Astra. Astra still leads at 68.1%, and OpenAI recommends it for the hardest research. Factual error rates at low effort fall from 11.4% to 7.7%, a reduction of about 32%.
Developers get a 1,050,000-token context window, 128,000 max output tokens, and an April 30, 2026 knowledge cutoff, with text and image input and text output. reasoning.effort accepts low, medium (default), high, xhigh, and max, while the Responses API is required for tool calling. Prompts above 272K input tokens cost 2x input and cache rates plus 1.5x output; Batch and Flex are 50% cheaper, Fast mode costs 2x, and fine-tuning is unsupported.
Competitively, Claude Sonnet 5.5 matches Sol's $2 and $10 list prices but charges twice as much for cached input, while Gemini 3.1 Pro charges $12 for output and stays in preview. OpenAI plans a GPT-6.1 Sol Ultrafast option in Codex within days, promising up to 8x faster token generation.
