Key details

  1. GPT-6 Sol and GPT-6 Luna launched September 22, 2026.
  2. Sol API pricing is $2/M input and $10/M output.
  3. Luna API pricing is $0.10/M input and $0.50/M output.
  4. OpenAI says both are 50% cheaper than their GPT-5.6 promotional pricing.
  5. OpenAI reports Sol at 68.8% on DeepSWE v1.1 at max effort.
  6. OpenAI reports Luna at 66.6% on DeepSWE v1.1 at max effort.
  7. Cached GPT-6 input reads receive a 90% discount.
  8. Reasoning-effort and tool-availability changes can preserve earlier cached context.
  9. The API model IDs are `gpt-6-sol` and `gpt-6-luna`.
  10. The models launched in ChatGPT Work and Codex for eligible paid users; Luna is also available to Free and Go users in the desktop app.

What builders should take away

  1. Re-run model routing economics: tasks previously sent to GPT-5.6 Sol or a more expensive frontier model may now fit GPT-6 Luna or Sol.
  2. Measure cost per accepted task, including retries and tool calls, rather than comparing only per-token rates.
  3. Instrument cache-hit rates for long-running agents; a 90% cached-input discount can matter as much as the base model price.
  4. Keep Astra as an escalation path for the hardest tasks instead of assuming every workflow benefits from the flagship.
  5. Reproduce OpenAI's coding and agent benchmarks on your own repositories because vendor task-cost comparisons depend heavily on harness and effort settings.

What changed

OpenAI released GPT-6 Sol and GPT-6 Luna on September 22, 2026. API list prices are $2/M input and $10/M output for Sol and $0.10/M input and $0.50/M output for Luna, which OpenAI describes as 50% below GPT-5.6 promotional pricing. The models inherit methods from GPT-6 Astra and target cheaper professional, coding, agent and computer-use workloads. OpenAI also improved GPT-6 prompt caching: cached input reads receive a 90% discount, reasoning-effort and tool-availability changes can preserve earlier cached context, and developers get cache diagnostics and explicit prefix breakpoints.

Why it matters

For builders running agents continuously, model economics are moving from token sticker price toward cost per completed task and cost of repeatedly reprocessing context. Sol and Luna lower both sides of that equation: list prices fall sharply while caching makes long-lived conversations and agent prefixes cheaper to reuse. This can change routing decisions even where the flagship Astra remains more capable, because many recurring coding, extraction, workflow and computer-use jobs do not need the most expensive tier.

OpenAI cut the recurring price floor, not just a launch promotion

GPT-6 Sol costs $2/M input and $10/M output; GPT-6 Luna costs $0.10/M input and $0.50/M output. OpenAI says those are 50% below the GPT-5.6 promotional rates. Independent specialist coverage characterizes the models as broadly preserving predecessor-class performance while making price the larger generational change.

The coding economics are aimed directly at sustained agent use

OpenAI reports GPT-6 Sol at 68.8% on DeepSWE v1.1, within 1.1 percentage points of the Claude Fable 5 result it cites while costing about 80% less per task under its assumptions. Luna reaches 66.6% at max effort and is claimed to cost 93% less per task than Opus 5 and 96% less than Fable 5 in the cited comparison. These are OpenAI-presented comparisons and should be reproduced on real repositories.

Caching becomes part of the agent cost model

GPT-6 cached input reads are discounted by 90%. OpenAI says changing reasoning effort or enabled tools can now preserve earlier context for cache reuse, and explicit cache breakpoints plus diagnostics make missed caching visible. GitHub reports that related improvements reduced the share of prompt tokens requiring fresh processing by more than 50% across billions of OpenAI-model requests in Copilot.

Astra stays the escalation tier

OpenAI still positions GPT-6 Astra as its best model. The practical architecture is therefore increasingly tiered: Luna for high-volume inexpensive work, Sol for stronger recurring coding and agent tasks, and Astra for the hardest cases. Applications that already route by task difficulty can potentially capture the new economics without changing their overall orchestration design.

What to watch next

  • Independent production comparisons of GPT-6 Sol and Luna against Opus 5.5, Fable 5.1 and MiMo-V2.6.
  • Whether OpenAI changes or expands Batch, Flex and Fast pricing for the new models.
  • Real cache-hit rates in long-running coding agents and multi-tool workflows.
  • Whether Sol or Luna replace older models as defaults in Codex and third-party routing platforms.
  • How the lower prices affect inference-provider and open-weight model pricing.

Still unclear

  • Most launch benchmark and task-cost comparisons are OpenAI-produced.
  • The 50% comparison is against GPT-5.6 promotional pricing, not necessarily every historical or provider-specific rate.
  • Cache savings depend on application prompt structure and actual cache-hit rates.
  • Production ChatGPT behavior can differ from API evaluations because tools, prompts and orchestration differ.

Sources

Direct reading behind this dossier.

2 sources
Introducing GPT-6 Sol and Luna
OpenAI primary

Primary launch source for pricing, benchmarks, caching changes, availability and model positioning.

Discussion

Discussion is reader-contributed. Comments are not part of the BTN dossier or its editorial evidence.

0 visible comments

Join the discussion

Keep comments useful and relevant. Reader contributions may be moderated and are not BTN editorial evidence.

Sign in to comment