October 26 update: where GPT-6.1 Terra fits
OpenAI released GPT-6.1 Terra on October 26, 2026. This adds another option to the Sol/Luna buying decision; it does not replace the launch history or make older model IDs interchangeable. OpenAI positions Terra for everyday professional work, visual reasoning and computer use, while recommending GPT-6.1 Sol and GPT-6 Astra for the most complex coding workloads. These are vendor positioning statements, not our benchmark results. See the official launch announcement.
The API ID is gpt-6.1-terra. At Standard rates for requests with at most 272K input tokens, USD per million tokens is $1.00 input, $0.05 cached input and $5.00 output; cache writes cost $1.25 for 30 minutes or $2.50 for 12 hours. Above 272K input tokens, input, cached-input and cache-write rates double and output rates increase by 50% across the entire request. The model supports a 1,050,000-token context and up to 128,000 output tokens; tool calling requires the Responses API. Check the model documentation for processing-mode and regional-pricing conditions.
Separately, GitHub announced a gradual Copilot rollout for Pro, Pro+, Max, Business and Enterprise. Organization policy can affect availability. This is not proof that the model is already selectable in every account, and the direct OpenAI API rates above are not a verified Copilot price quote.
Our assessment: Terra is worth evaluating where image understanding or computer interaction matters and API unit cost constrains throughput. Do not migrate a demanding coding workflow solely because its listed token rates are lower: compare task completion, retries, latency and total billed usage on your own representative workload. Existing Sol/Luna integrations can remain the baseline until those checks demonstrate a useful trade-off. Independent workload results, stable account availability and clarified platform billing would change this recommendation; we have not run that comparison.
OpenAI launched GPT-6 Sol and GPT-6 Luna on September 22, 2026, expanding the GPT-6 family beyond Astra with two cheaper models aimed at everyday professional work, coding, agents, and high-volume API workloads. The headline change is price: OpenAI lowered Sol and Luna API prices compared with the promotional pricing of their GPT-5.6 predecessors.
The important part is not just that two new model names appeared. GPT-6 now has a clearer cost-performance ladder: Astra for maximum capability, Sol for demanding work at a much lower cost, and Luna for fast, high-volume workloads where unit economics matter most.
September 29 update: what GPT-6.1 Sol changes for API buyers
OpenAI released GPT-6.1 Sol on September 29, 2026. The API identifier is gpt-6.1-sol. Standard input and output prices stay at GPT-6 Sol’s levels; cached reads fall from $0.20 to $0.10 per million tokens. The September 22 Sol/Luna launch and September 25 image-fix history remain below. See the September 29 API changelog.
GPT-6.1 Sol pricing: check request size before comparing costs
USD per million tokens, Standard processing. Short context means at most 272K input tokens; long context means more than 272K. Once that threshold is crossed, the long-context rates apply to the entire request, not just the excess tokens.
| GPT-6.1 Sol request | Input | Cached input | Cache write (30-minute default) | Output |
|---|---|---|---|---|
| Short: input ≤272K | $2.00 | $0.10 | $2.50 | $10.00 |
| Long: input >272K | $4.00 | $0.20 | $5.00 | $15.00 |
Batch and Flex cost 50% less than Standard; Fast costs twice Standard. Eligible regional processing adds 10%. These are token rates, not a promise about total task cost. See OpenAI API pricing.
Migration checks and access
The model reference lists a 1,050,000-token context window, 128,000 maximum output tokens, and an April 30, 2026 knowledge cutoff. Use Responses for tool calling; Chat Completions supports this model without tool calling. Supported reasoning efforts are low, medium (default), high, xhigh, and max; none and minimal are unsupported.
OpenAI’s launch announcement lists access through ChatGPT Work and Codex for Plus, Pro, Business, Enterprise, and Edu, plus the API. It explicitly says GPT-6.1 Sol is not yet available in Chat. Sol Ultrafast was announced as coming in the following days; that is separate from Astra Ultrafast’s September 29 API release.
OpenAI reports near-Astra results on selected coding, computer-use, and professional-work evaluations. These are vendor results, not AI-XBlog tests or a guarantee of production quality, latency, or savings. Those outcomes remain unverified here. For migration, compare GPT-6 Sol and GPT-6.1 Sol on your own prompts and tools, recording successful-task cost and cache reuse before changing your default.
September 22 launch reference: GPT-6 Sol and Luna at a glance
| Model | API input | API output | API model ID | Best fit |
|---|---|---|---|---|
| GPT-6 Astra | Premium tier | Premium tier | gpt-6-astra |
Maximum capability and highest-stakes work |
| GPT-6 Sol | $2 / 1M tokens | $10 / 1M tokens | gpt-6-sol |
Complex professional work, coding, agents, computer use |
| GPT-6 Luna | $0.10 / 1M tokens | $0.50 / 1M tokens | gpt-6-luna |
High-volume, cost-sensitive tasks and background automation |
Prices above are OpenAI standard API prices announced for Sol and Luna. Astra remains the premium model in the family.
Verified model contract
For citation, procurement and implementation checks, the table below records the current provider-documented contract for GPT-6 Sol and Luna as verified on September 24, 2026. These are OpenAI documentation values, not AI-XBlog benchmark results.
| Field | GPT-6 Sol | GPT-6 Luna |
|---|---|---|
| API model ID | gpt-6-sol |
gpt-6-luna |
| Context window | 1,050,000 tokens | 1,050,000 tokens |
| Maximum output | 128,000 tokens | 128,000 tokens |
| Knowledge cutoff | April 20, 2026 | May 18, 2026 |
| Standard input / 1M tokens | $2.00 | $0.10 |
| Cached input / 1M tokens | $0.20 | $0.01 |
| Cache write / 1M tokens (30-minute default) | $2.50 | $0.125 |
| Standard output / 1M tokens | $10.00 | $0.50 |
Long-context and processing note: OpenAI states that prompts above 272K input tokens are charged at 2× input/cache rates and 1.5× output for the full request. Batch and Flex are priced at 50% of Standard, Fast mode at 2× applicable rates, and eligible regional processing carries a 10% uplift. EU data residency for Sol and Luna is available with Standard, Fast, Flex, and Batch processing (official documentation rechecked October 9, 2026). The data-residency eligibility table specifically confirms Fast mode for Responses and Chat Completions; availability still depends on the eligible endpoint and project. See the GPT-6 Sol model page, GPT-6 Luna model page, and OpenAI API pricing.
What changed from GPT-5.6?
The clearest change is the API price cut. OpenAI says GPT-6 Sol drops from GPT-5.6 Sol’s promotional $4 input / $20 output per million tokens to $2 / $10. GPT-6 Luna drops from $0.20 / $1.20 to $0.10 / $0.50.
- Sol input: 50% cheaper than GPT-5.6 Sol promotional pricing.
- Sol output: 50% cheaper.
- Luna input: 50% cheaper.
- Luna output: falls from $1.20 to $0.50 per million tokens.
- Model family: both inherit training advances introduced with GPT-6 Astra, according to OpenAI.
OpenAI also says Sol makes about half as many mistakes as its predecessor on an internal factuality evaluation based on de-identified conversations that previously triggered user-reported factual errors. That is an OpenAI evaluation, not an independent benchmark, so it should be treated as a vendor claim rather than a universal error-rate guarantee.
GPT-6 Sol vs Luna vs Astra: which should you use?
Choose GPT-6 Sol when the task is expensive to get wrong
Sol is the most interesting release for developers and teams that previously wanted near-frontier capability but could not justify Astra-level cost for every request. OpenAI positions it for difficult professional work, coding, computer use, and agentic workflows.
Choose GPT-6 Luna when throughput matters more than maximum reasoning depth
At $0.10 per million input tokens and $0.50 per million output tokens, Luna changes the economics of classification, routing, extraction, lightweight coding assistance, repetitive background agents, and other workloads where millions of calls can matter more than squeezing out the final increment of capability.
Choose GPT-6 Astra when you need the strongest model
OpenAI still describes Astra as its best model across the board. If a workflow is unusually difficult, high stakes, or worth materially more than the inference cost, Astra remains the top-end option.
Availability: ChatGPT Chat, Work, Codex, desktop, and API
October 7, 2026 update: GPT-6 and Intelligent UI in Chat
OpenAI’s October 7 announcement starts a global, gradual Chat rollout: GPT-6 Sol for Plus, Pro, Business and Enterprise, followed by GPT-6 Luna for Free and Go starting October 8. These are rollout start dates, not confirmation that every eligible account has access. Enterprise permissions still apply; no precise country-by-country activation time is given.
Intelligent UI can turn an answer into a useful mix of text, visuals and interactive controls, while ordinary text answers remain possible. The ChatGPT release notes place it at Instant through Extra High, subject to plan access. Pro reasoning still uses GPT-6 Astra and does not support Intelligent UI; that option is different from having a Pro subscription. This Chat release does not change the models powering Work or Codex.
The current model guide gives Plus access to Medium and High, but not Extra High or Pro. Existing Latest selections move to GPT-6 with their effort preserved; an explicit GPT-5.6 Sol selection stays on GPT-5.6. At our October 7, 2026, 19:31 UTC check, the guide still described Free and Go as using GPT-5.6 Luna, so the announced October 8 expansion should not be read as completed account migration.
Use the web or a supported updated app. The older ChatGPT macOS and Windows desktop apps do not support these new capabilities. Workspace permissions and staged client availability can affect what appears.
For API readers, the October 7 API changelog separately records an update to chat-latest. That entry does not announce an API for ChatGPT’s Intelligent UI renderer. It is not a reason to replace the dated GPT-6.1 Sol pricing and model-contract analysis above.
Our assessment: check the plan, selected effort and supported client before treating a missing interface as a model failure. Evaluate task quality separately: a new display format alone does not establish greater accuracy or a lower API bill. We have not tested account access or the new interface ourselves.
September 22–23 launch history: Work, Codex and desktop
At the September 22 launch, OpenAI announced GPT-6 Sol and Luna in ChatGPT Work and Codex for Plus, Pro, Business, Enterprise and Edu users, plus GPT-6 Luna access for Free and Go users in the desktop app. That historical desktop access should not be confused with the October Chat and Intelligent UI rollout.
Codex CLI update: OpenAI released Codex CLI 0.156.1 on September 23 with GPT-6 Sol and GPT-6 Luna added directly to the model picker. The CLI’s rate-limit switch prompt now recommends GPT-6 Luna, giving Codex users an explicit lower-cost fallback path when usage limits become the constraint.
At the September 22 launch, OpenAI said the models were not yet available in regular Chat, and described a gradual Work and Codex rollout. The dated October 7 update above supersedes that launch-time Chat restriction for GPT-6 Sol and Luna; GPT-6.1 Sol is a separate Work, Codex and API release.
For developers, the API model IDs are straightforward: gpt-6-sol and gpt-6-luna.
Microsoft Foundry availability
Microsoft says GPT-6 Sol and GPT-6 Luna are also generally available in Microsoft Foundry alongside GPT-6 Astra. Standard deployment is available across all 28 Global regions plus U.S. and EU Data Zones. Provisioned Throughput is available for Astra and Sol across Global, U.S. and EU Data Zones, while Priority Processing for Sol is available across Global and U.S. Data Zones.
For Azure buyers, deployment choice changes the bill. Microsoft lists Global Standard short-context pricing at $2 input / $10 output per 1M tokens for Sol and $0.10 / $0.50 for Luna, matching OpenAI’s headline standard rates. Long-context Global Standard rises to $4 / $15 for Sol and $0.20 / $0.75 for Luna. Microsoft lists U.S. Data Zone pricing at a 10% premium to Global and EU Data Zone pricing at a 20% premium, so teams should not assume one GPT-6 price applies to every Azure deployment.
Prompt caching is a bigger part of the cost story
Prompt-cache pricing checked October 21, 2026 (Asia/Taipei): For GPT-5.6 and later, including Sol and Luna, the default 30-minute cache lifetime costs 1.25× uncached input for writes. Following OpenAI’s October 20 announcement, Responses API requests can select prompt_cache_options.ttl: "12h", with writes billed at 2.5× instead. Both multipliers are total write rates, not extra surcharges. The table write prices above describe the default 30-minute option. Read rates do not change with lifetime: Sol and Luna reads cost 10% of uncached input; GPT-6.1 Sol reads cost 5%. Twelve hours is a minimum cache lifetime, not a deletion deadline or latency guarantee. See the October 20 changelog and prompt-caching guide.
Our assessment: choose the longer lifetime for stable prefixes reused across hours only when the expected read savings justify its higher write cost. For frequently changing context, retain the default and compare actual billed writes, cache reads and successful-task cost before switching. We have not benchmarked this change.
This matters because the cheapest model is not always the cheapest workflow. Long-running agents repeatedly sending the same instructions, repository context, policies, or tool definitions can spend a large share of their budget reprocessing context. Better cache reuse can reduce that cost independently of the headline token rate.
What OpenAI’s benchmark claims do — and do not — prove
OpenAI published strong results for Sol and Luna across AutomationBench, FrontierCode, DeepSWE, OSWorld, and its internal factuality testing. For example, the company reports GPT-6 Sol at xhigh effort scoring 33.2% on AutomationBench at $0.27 per task, and says Sol approaches Astra-level factual reliability at much lower cost.
Those numbers are useful for understanding OpenAI’s intended positioning, but AI-XBlog is not treating vendor benchmark results as an independent head-to-head test. Production performance still depends on prompts, tool access, effort settings, workflow design, latency constraints, and the cost of failed tasks.
September 25, 2026: OpenAI fixed an image-encoding bug
OpenAI says it fixed an image-encoding bug in GPT-6 Sol and GPT-6 Luna on September 25, 2026 that had degraded image understanding. The fix applies to visual tasks in the API and Codex, including computer use. OpenAI recommends that teams using image inputs rerun their evaluations and retry workflows that may have been affected.
This was a model-quality fix, not a new model launch or pricing change. OpenAI’s September 25 API changelog did not announce new token rates alongside the fix, so the pricing documented above remains the relevant published rate unless OpenAI updates its pricing pages separately. For developers, the practical consequence is straightforward: if an evaluation, screenshot workflow, visual agent or computer-use task produced unexpectedly weak results before September 25, rerun it before treating the earlier result as representative of current GPT-6 Sol or Luna behavior.
Evidence: OpenAI API changelog, September 25, 2026. AI-XBlog has not independently benchmarked the pre-fix versus post-fix models, so any performance improvement beyond OpenAI’s stated fix remains unverified here.
The practical decision framework
October 14, 2026 API deprecation notice: OpenAI has scheduled gpt-4o-2024-08-06, gpt-4o-2024-11-20, gpt-4o-mini-2024-07-18, gpt-4.1-2025-04-14, and gpt-4.1-mini-2025-04-14 for removal from the API on January 14, 2027. Fine-tuned models based on those snapshots follow the same schedule. The recommended replacement is gpt-6-luna. This is a future shutdown of the listed API snapshots, not evidence of a current outage or a retirement date for every GPT-4 model or ChatGPT plan. See the official deprecations notice.
The GPT-6 Luna reference currently marks fine-tuning as unsupported. Luna supports function calling in Chat Completions only with reasoning_effort: "none"; use Responses for reasoning with tools. The migration guide also documents parameter and cache-setting differences.
Our assessment, not a migration benchmark: inventory the exact snapshot IDs in application configuration before assigning migration work. For fine-tuned deployments, budget for rebuilding and evaluating the behavior rather than assuming the fine-tune transfers. For tool-using applications, test the endpoint, reasoning setting, argument validity, output format and failure handling together. Compare cost per successful task on representative workloads before changing the default. The deadline makes this planning time-sensitive; it does not establish that a particular workflow is affected or that replacing the model string is sufficient.
- Start with Luna for cheap routing, extraction, triage, and repetitive background work.
- Move to Sol when failures, retries, coding quality, or long-horizon agent behavior create more cost than the model price itself.
- Escalate to Astra only when the task’s value justifies paying for the strongest available model.
- Measure cache hit rate before assuming token list prices tell the whole story.
- Route by task difficulty rather than hard-coding every request to one model.
For teams choosing a subscription rather than an API model, see AI-XBlog’s ChatGPT pricing guide. For production agent architecture, see the OpenAI Agents API guide, AI computer-use guide, and AI agent security guide.
FAQ
How much does GPT-6 Sol cost?
OpenAI’s launch pricing is $2 per million input tokens and $10 per million output tokens.
How much does GPT-6 Luna cost?
OpenAI’s launch pricing is $0.10 per million input tokens and $0.50 per million output tokens.
What are the GPT-6 Sol and Luna API model IDs?
The API identifiers are gpt-6-sol and gpt-6-luna.
Are GPT-6 Sol and Luna available in normal ChatGPT chats?
Chat access is rolling out. OpenAI announced a global start on October 7, 2026 for GPT-6 Sol on Plus, Pro, Business and Enterprise, with GPT-6 Luna for Free and Go starting October 8. Access is gradual and workspace permissions apply. The September 22 launch was limited to other surfaces; GPT-6.1 Sol remains a separate release. Check the dated availability section above for Intelligent UI, plan and client restrictions.
Sources
- OpenAI: Introducing GPT-6 Sol and Luna
- OpenAI: Better prompt caching for GPT-6
- OpenAI: GPT-6 Astra
- Reuters: OpenAI expands GPT-6 lineup with cheaper Sol and Luna models
- TechCrunch: OpenAI launches GPT-6 Sol and Luna
- OpenAI Codex CLI 0.156.1 release notes
- OpenAI: GPT-6 Sol model documentation
- OpenAI: GPT-6 Luna model documentation
- OpenAI: API pricing
- Microsoft Azure: GPT-6 Astra, Sol, and Luna for production agents in Microsoft Foundry
Chat and Intelligent UI update checked: October 7, 2026, against OpenAI’s announcement, ChatGPT release notes, current model guide and API changelog. Account-level rollout and interface behavior were not independently tested. GPT-6.1 Sol API reporting was checked September 30; earlier Sol/Luna and Microsoft Foundry reporting was checked September 28. Those historical checks are not a fresh audit of every price or deployment option.
AI-XBlog Weekly Brief
Keep up with AI that actually works
Join the AI-XBlog Weekly Brief for major AI updates, practical workflows, useful tools, and editor’s picks. No daily noise.
Reader discussion
Join the discussion
Have you tried this tool or workflow? Share your experience, corrections, or questions. Useful reader feedback may help us improve this article.
All comments are reviewed before publication. Your email address will not be published. Promotional links and low-value spam are removed.
