Business workflows
WinnerAutomationBench 1.0.6
Business workflows across 47 tools. Efforts differ as labeled. Fable 5.1 includes Opus 5 fallback on about 40% of tasks. Source: OpenAI.
SPIRITT
GPT-6 SolGPT-6 Sol. Bigger work. Smaller bills.
Tackle large code changes, longer research runs, and multi-step business work. GPT-6 Sol pairs capable coding and computer use with half the API token price of GPT-5.6 Sol, plus a 1.05M context window to keep more of your project in view.
GPT-6 Sol in action
See builders turn Sol into websites, explorable 3D scenes, games, videos, and real code changes, with firsthand comparisons along the way.
More room to build, test, and iterate
Sol is the middle ground for ambitious projects: more capable than the smallest model, with a much lighter bill than the flagship.
Complex work without the flagship price
OpenAI positions GPT-6 Sol for complex coding and agentic workflows. It supports text and image input, up to 922K input tokens within a 1.05M total context window, and up to 128K output tokens. Medium is the default reasoning effort, with options from none to max.
Standard API prices are $2/M input and $10/M output, both 50% below GPT-5.6 Sol. Cached input costs $0.20/M. Reusing project context can make a meaningful difference in long coding sessions and repeated agent work.
Independent testing gives the price cut practical context: Artificial Analysis reports $1.06 per Intelligence Index task versus $1.99 for GPT-5.6 Sol, both at max effort. Its Coding Agent Index rises from 55 to 57, while broad intelligence stays roughly level and some knowledge-work results regress.
Complex codingMulti-step agents1.05M context128K output$2 / $10 per M90% cached-input discount
AA Intelligence Index v4.3
Artificial Analysis, September 22. All rows use max effort; the two Claude models include fallback. A broad capability comparison, not a task-price ranking.
What Sol delivers for the cost
Compare code changes, business workflows, computer use, and the cost of keeping an agent working. Effort levels and evaluation setups are labeled so like-for-like comparisons stay clear.
Workflow economics
WinnerAutomationBench cost per task
Same settings as the score card. Fable’s shown cost excludes Opus 5 fallback; its actual total is higher. Source: OpenAI.
Long-running work
Agents’ Last Exam V1
Long-horizon professional workflows. Opus 5 uses high, its best evaluated setting in this source; the OpenAI rows use max. Source: OpenAI.
Merge-ready code
FrontierCode 1.1 Main
Changes are judged for correctness and merge-readiness. Fable 5.1 medium is shown rather than a weaker higher-effort run. Source: OpenAI.
Software engineering
DeepSWE v1.1: OpenAI evaluation
Complex repository work. Efforts are labeled; these official research/API runs differ from Artificial Analysis’s Codex evaluation below.
Computer use
OSWorld 2.0 offline
Partial reward on the v2026.08.08 offline set, not full-task completion. Both generations are shown at max; any other effort is labeled. Source: OpenAI.
Factual reliability
Factual error rate on difficult prompts
Lower is better. Prompts were selected from user-flagged mistakes, so these are not ordinary-usage error rates. Max effort. Source: OpenAI.
Independent coding
AA Coding Agent Index v1.5
Artificial Analysis, September 22. Codex for OpenAI; Claude Code for Claude, with fallback for Fable. All rows use max effort.
Independent repository work
WinnerTerminal-Bench 4.0 in Codex
Artificial Analysis’s Coding Agent evaluation at max effort. Same Codex setup for both generations; separate from the official OpenAI comparison.
Independent code understanding
WinnerSWE-Atlas-QnA in Codex
Repository questions in the same Codex evaluation at max effort. Published in Artificial Analysis’s September 22 launch report.
Independent cost comparison
WinnerAA Intelligence Index cost per task
September 22 v4.3 launch snapshot, max effort. The models use more output tokens than their predecessors; the savings are driven by lower prices.
Knowledge reliability
WinnerAA-Omniscience hallucination rate
Lower is better. AA’s special hallucination metric is not an all-response error rate. Accuracy is shown separately; declining questions can change both.
Knowledge trade-off
AA-Omniscience accuracy
Correct answers across all questions at max effort. Sol answers fewer questions, reducing hallucinations but also lowering accuracy. Source: Artificial Analysis.
Communication and trust
Coding-deception rate
Lower is better. OpenAI deliberately selected difficult, dishonesty-inducing coding tasks. Max effort; not a typical usage rate.
Tool-use transparency
Failure to disclose broken search
Lower is better. This challenge tests whether an agent admits that its search tool is broken. Max effort; not a typical usage rate.
Published API pricing
WinnerInput price per million tokens
OpenAI Standard rates for requests up to 272K input tokens. Both generations’ input price drops by 50%.
Published API pricing
WinnerOutput price per million tokens
OpenAI Standard token rates. Sol’s output price drops by 50%; total task costs depend on token use and settings.
Published API pricing
WinnerCache-read price per million tokens
Cached reads cost 10% of uncached input. Repeated project context can use the discount when cache requirements are met.
Independent output speed
Output tokens per second
Artificial Analysis snapshot, September 23, OpenAI API at max effort. Speed after the first chunk, not a guarantee of total task time.
Sources: OpenAI’s September 22 launch and GPT-6 system-card appendix, official API documentation, and Artificial Analysis. Each chart keeps its evaluator, version, effort, and fallback conditions. OpenAI’s comparator runs come from public reports; Fable 5 appears where 5.1 results were unavailable. The AA Intelligence and Coding Agent launch snapshots use v4.3 and v1.5 respectively. Prices are OpenAI API list rates, not SPIRITT plan prices; long-context, tool, and optional service charges can apply. Output speed is a separate September 23 observation.
How It Works
Give the work a home, choose the right model, and keep the result moving.
01
Bring the work into SPIRITT
Start with your files, repository, or business goal. A workspace brings your agent, browser, terminal, integrations, and memory together around the project.
02
Match the model to the job
Check the models and connection options enabled in your workspace. Choose the level of capability and cost that fits the work, then set a clear finish line.
03
Check the result, then keep going
Work toward a usable output: a tested change, a finished report, or a repeatable workflow. Keep the files and decisions together, and review important actions before they go live.
Put your next project to work
Bring your agent the tools, context, and workspace to turn a request into something useful.
Questions
Frequently asked questions
01What is GPT-6 Sol best suited for?+
Complex coding, research, and multi-step agent workflows where you need room to iterate without paying flagship token prices on every step. OpenAI positions Sol between the higher-capability Astra and the lower-cost, high-volume Luna.
02How much cheaper is it than GPT-5.6 Sol?+
Standard input pricing falls from $4 to $2 per million tokens, and output from $20 to $10: a 50% reduction in each. Cached input is $0.20/M and cache writes are $2.50/M. Total task cost still depends on tokens, reasoning effort, tools, and processing mode. These are OpenAI API list prices, not SPIRITT plan prices.
03Is Sol better at every task?+
No. Artificial Analysis reports better Coding Agent Index results at lower cost, but roughly level overall intelligence. It also found lower GDPval-AA results. On AA-Omniscience, Sol hallucinates less partly by declining more questions; accuracy falls from 59% to 54% in that evaluation. Review outputs against what the task actually requires.
04When should I choose Luna or Astra instead?+
Luna is the lower-cost choice for focused, repetitive work and high request volumes. OpenAI positions Astra as its highest-capability model for the hardest tasks. Sol balances capable project work with a lower running cost.
05How much context can it use?+
The documented total context window is 1,050,000 tokens, with a maximum of 922,000 input tokens and 128,000 output tokens. A large window helps keep more material together; it does not guarantee that every detail will be used correctly.
06Do long prompts cost extra?+
Yes. Above 272K input tokens, the entire request is billed at 2x the normal input and cache rates and 1.5x the output rate. Tool calls and optional processing settings can add charges too.
07What do Batch, Flex, and Fast mode change?+
Batch and Flex use 50% of Standard token rates, with different latency and availability trade-offs. Fast mode costs 2x the applicable rates in exchange for higher throughput; it is not a guarantee of a particular total task speed. Regional processing adds 10% where available, and EU data residency requires Standard processing.
08Can it create images or use the computer?+
Its native inputs are text and images, and its native output is text. Through the Responses API it can use tools for image generation, web and file search, code execution, hosted shell, computer use, and more. Visual demos use those tools; they are not evidence of native audio or video output.
09Can I use GPT-6 Sol in my SPIRITT Workspace?+
Check the models and connected options enabled in your workspace. Access to a newly released model depends on the available connection and workspace setup; its OpenAI release does not automatically enable it in every workspace.
Buy from builders who use what they sellBuilt using