Paid⚡ Featuredchat

GPT-5.6

Sol, Terra, and Luna access and pricing tracker after GPT-5.6's public release.

OpenAI's GPT-5.6 model family status page for Sol, Terra, and Luna access tracking, current API pricing, ChatGPT plan availability, Work/Codex/API access, and safety notes. GPT-5.6 Sol is listed at $4 per 1M input tokens and $20 per 1M output tokens through at least November 21, 2026.

4.7

Try GPT-5.6 Free →

Opens official website

openaigpt-5.6model accesssol terra luna

Live change feed

Recent verified changes

Shared with the homepage and tool pages

pricinggpt-5-6Aug 21, 2026

OpenAI cuts GPT-5.6 Sol API pricing by 20% for input and 33% for output

GPT-5.6 Sol now costs $4 per 1M input tokens and $20 per 1M output tokens. OpenAI describes this as promotional pricing available at least through November 21, 2026; subscription prices and quota budgets are unchanged.

Verify official source →

pricinggpt-5-6Jul 30, 2026

OpenAI cuts GPT-5.6 Luna API prices by 80%

Luna now costs $0.20 per 1M input tokens and $1.20 per 1M output tokens. Subscription prices and quota budgets are unchanged, while Luna and Terra consume fewer credits in ChatGPT Work and Codex.

Verify official source →

verifiedLast verified 2026-08-14

Luna price falls 80% as GPT-5.6 targets high-volume work

OpenAI reduced GPT-5.6 Luna API pricing to $0.20 per 1M input tokens and $1.20 per 1M output tokens, an 80% cut. Terra now costs $2 input and $12 output per 1M tokens. Subscription prices and quota budgets remain unchanged.

Luna API: $0.20 input / $1.20 output per 1M tokens
Terra API: $2 input / $12 output per 1M tokens
Subscription impact: Same prices and quota budgets; fewer credits consumed
Availability: ChatGPT Work, Codex, and API

Verify source: OpenAI price-performance announcement →

About GPT-5.6

GPT-5.6 is OpenAI's released July 2026 model family for more capable reasoning, coding, agentic work, cybersecurity assistance, and long-running product workflows. The family includes GPT-5.6 Sol, the flagship model; GPT-5.6 Terra, the balanced lower-cost option; and GPT-5.6 Luna, the fastest and lowest-cost option. Standard ChatGPT conversations expose Sol through eligible reasoning options, while Terra and Luna are available through Work in ChatGPT, Codex, and the OpenAI API rather than as standard ChatGPT picker options. The API now supports Programmatic Tool Calling, explicit prompt caching controls, persisted reasoning across turns, configurable reasoning effort (including a max level), Pro mode, and a Responses API multi-agent orchestration beta. The gpt-5.6 model alias routes to gpt-5.6-sol by default. AI Linkbase tracks GPT-5.6 as a released model family with product-specific access rules, safety caveats, and ranking review flags.

✅ Pros

+Full API support: Responses API, Chat Completions, and Batch API across all three models
+Programmatic Tool Calling enables structured, reliable tool use in production pipelines
+Explicit prompt caching controls reduce cost on repeated context windows
+Persisted reasoning maintains chain-of-thought state across multi-turn conversations
+Configurable reasoning effort (including max level) lets developers trade speed for depth
+Responses API multi-agent orchestration beta supports complex agent workflows
+Sol, Terra, and Luna create a useful high-end to lower-cost segmentation
+Worth monitoring for ChatGPT, API, and Codex users planning model upgrades

❌ Cons

−Pricing, rate limits, and plan limits still need workflow-specific verification
−Multi-agent orchestration is still in beta and may change
−Regional and enterprise policy restrictions may change during rollout
−Ranking changes should wait for benchmarks, pricing, and real access data

Key Features

Sol, Terra, and Luna Tracking

Tracks OpenAI's GPT-5.6 family as separate versions so users do not confuse a preview or staged rollout with universal availability.

Programmatic Tool Calling

Structured tool-use API lets developers define tools and receive deterministic function calls, replacing brittle prompt-based extraction.

Prompt Caching Controls

Explicit caching API reduces cost on repeated system prompts and context windows across requests.

Persisted Reasoning

Chain-of-thought state can persist across turns, enabling multi-step workflows without re-deriving context.

Multi-Agent Orchestration Beta

Responses API beta supports spawning and coordinating multiple agents for complex task decomposition.

Configurable Reasoning Effort

Developers can set reasoning effort from low to max, trading latency for depth on a per-request basis.

Pricing Plans

ChatGPTPlan-based

✓GPT-5.6 Sol in eligible reasoning options
✓Free and Go plan limitations apply
✓Terra and Luna not selectable in standard conversations

Luna API$0.20input / $1.20 output per 1M tokens

✓80% lower API price from July 30
✓Fastest and lowest-cost GPT-5.6 option
✓Designed for high-volume work

Terra API$2input / $12 output per 1M tokens

✓20% lower API price from July 30
✓Balanced everyday-work model
✓Available in Work, Codex, and API

Prices, plan limits, and promotions may change. Verify final pricing on the official site. Last verified: 2026-08-14.

🏆 Our Verdict

GPT-5.6 is a released OpenAI model family with serious API capabilities — Programmatic Tool Calling, prompt caching, persisted reasoning, and multi-agent orchestration bring it closer to production-grade agent infrastructure. Sol, Terra, and Luna do not mean the same thing across ChatGPT, Work, Codex, and the API, so teams should verify plan limits, pricing, and safety policies before migrating. The gpt-5.6 alias defaults to Sol; developers targeting cost efficiency should explicitly specify Terra or Luna.