GPT 6.1 Sol Nearly Matches Astra At One Fifth Price
OpenAI has introduced GPT 6.1 Sol, an upgrade to GPT 6 Sol that the company says nearly matches GPT 6 Astra's intelligence on agentic coding, computer use and professional work at one fifth of Astra's standard input and output token prices. Cached input is priced at ten cents per million tokens, which OpenAI describes as ninety five percent below its standard input pricing and fifty percent below GPT 6 Sol's cached input pricing.
Transcript
OpenAI has released GPT 6.1 Sol. It nearly matches GPT 6 Astra's intelligence at a much lower price.
GPT 6.1 Sol handles agentic coding and professional work. OpenAI charges one fifth of what that stronger model costs per token.
On real coding tasks it beats GPT 6 Sol by over six percentage points. On scientific workflows it more than doubles the earlier score.
It reaches paid users today in ChatGPT Work and Codex. Cached input costs ten cents per million tokens.
These are OpenAI's own numbers. The company says GPT 6 Astra still scores highest on the hardest scientific research and should be used there.
Inboxsmith builds AI receptionists for small business. Please like and subscribe for more news.
Sources
Every claim in this video comes from the top ranking coverage of this topic. The claims and where each one came from:
- "We're introducing GPT-6.1 Sol, an upgrade to GPT-6 Sol that nearly matches GPT-6 Astra's intelligence on agentic coding, computer use, and professional work." Backs the hook's "nearly matches GPT 6 Astra's intelligence".(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)
- Subtitle: "Near-Astra intelligence for a fifth of the price." And: the upgrade lands "at one-fifth of Astra's standard input and output token prices." Backs the hook's "much lower price" and scene one's "one fifth of what that stronger model costs per token".(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)
- "GPT-6.1 Sol offers a new balance of capability and cost for important everyday work. It delivers substantial improvements over GPT-6 Sol across complex professional tasks, from writing and debugging code to understanding documents and executing multi-step business workflows." Backs scene one's "handles agentic coding and professional work".(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)
- "On DeepSWE v1.1, which evaluates complex software-engineering tasks in real codebases, GPT-6.1 Sol matches GPT-6 Astra at roughly one-fifth of the cost, while eclipsing GPT-6 Sol's best score by 6.4 percentage points at a lower reasoning effort and cost." Backs scene two's "on real coding tasks it beats GPT 6 Sol by over six percentage points".(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)
- "On Terminal-Bench Science 0.1, which evaluates scientific workflows including data analysis, simulation, and theorem proving, GPT-6.1 Sol more than doubles GPT-6 Sol's score at maximum reasoning effort at less than half the cost per task." Backs scene two's "on scientific workflows it more than doubles the earlier score".(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)
- "GPT-6.1 Sol is available starting today to all Plus, Pro, Business, Enterprise, and Edu users in ChatGPT Work and Codex." Backs scene three's "it reaches paid users today in ChatGPT Work and Codex".(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)
- "Its standard API prices are $2 per million input tokens, $0.10 per million cached input tokens, and $10 per million output tokens." And: "Cached input costs just $0.10 per million tokens." Backs scene three's "cached input costs ten cents per million tokens".(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)
- Every benchmark figure quoted above is reported by OpenAI about its own model in its own announcement, in first person: "On several of these evaluations, it approaches GPT-6 Astra's performance at substantially lower cost." No independent testing is cited. Backs scene four's "these are OpenAI's own numbers".(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)
- "GPT-6 Astra still achieves the highest score among the models tested at 68.1%, and should be used for the most difficult scientific research tasks." Backs scene four's second sentence.(OpenAI (openai.com), first party announcement titled "Introducing GPT-6.1 Sol", dated 2026.09.29, live URL returned HTTP 403 so the alert time snapshot captured 2026-09-29T18:05:14.717Z was used as primary.)