Modelgeneral1mo ago

GPT-5.6 review

OpenAI's next flagship after GPT-5 — announced June 26, 2026 as a preview and generally available since July 9, 2026 in three tiers (Sol, Terra, Luna) with published per-token pricing.

By Ravi Menon · Models & Benchmarks EditorVerified 2026-07-21
Maker
OpenAI
Launched
Jun 26, 2026
Pricing
paid
Visit official site
Firstlook

Our verdict

GPT-5.6 is a meaningful generational step from OpenAI, and the Sol/Terra/Luna tier structure is the real story: capability-cost transparency baked into the model name, with a clean 5x price spread from Luna to Sol and a 1.05M-token context window across the line. It went from June 26 preview to general availability on July 9, so anyone can buy it today. We have not run it ourselves yet, so there's no score here and no claim about how it performs — what we can say is that the pricing and access questions that made this a wait-and-see model at announcement are now answered.

First look — our read from the docs and sources below; not yet hands-on tested.

OpenAI announced GPT-5.6 on June 26, 2026 — and the developer community was clearly ready for it. It's the first meaningful flagship release since GPT-5, and it comes with a structural change in how OpenAI names its models: three tiers called Sol, Terra, and Luna, each targeting a different capability-cost operating point.

That naming choice is more interesting than it sounds. The GPT-4 era gave us a confusing sprawl — GPT-4, GPT-4 Turbo, GPT-4o, GPT-4o mini, o1, o3, o3-mini — and users regularly had to dig into the docs to figure out which one to actually use. Sol/Terra/Luna is OpenAI's apparent attempt to make the tradeoff explicit from the name.

The June 26 announcement was a preview. That part is over: since July 9, 2026 all six variants — gpt-5.6-sol, sol-pro, terra, terra-pro, luna, luna-pro — have been publicly purchasable, with real published per-token pricing. This page is still a first look rather than a hands-on review, because we haven't run it on our own workloads yet — but the access and pricing unknowns are now resolved.

What we know

Three tiers under the GPT-5.6 name, each with a -pro variant, for six models you can call today. Closed source. A 1,050,000-token context window and a 128,000-token output cap across the line, with text, image, and file input returning text.

Pricing per 1M tokens (input / output):

  • Sol — $5.00 / $30.00
  • Terra — $2.50 / $15.00
  • Luna — $1.00 / $6.00

The -pro variants are listed at the same price as their base tier.

What we still don't know

Benchmark numbers — OpenAI hasn't published comparative results against GPT-5 or across the tiers. And what the -pro variants actually change, given they're priced identically to their base models. Fine-tuning options are likewise unannounced.

Who this is for

Developers already in the OpenAI ecosystem, and anyone who wants a very large context window with multimodal input. The tier split is the practical decision: Luna for high-volume work where $1/$6 per 1M tokens matters, Terra as the sensible default, Sol when you want the ceiling and can absorb $30 per 1M output tokens. The 5x spread from Luna to Sol is wide enough that benchmarking your own workload across all three is worth the afternoon.

The honest take

GPT-5.6 has OpenAI's full flagship backing, a structural naming update that's long overdue, and — now — public pricing and open access. The Anthropic flagship Claude Mythos landed the same week and remains the obvious head-to-head. What's missing is performance data: no public benchmarks from OpenAI, and no hands-on numbers from us.

This page will get a real verdict — including a score — once we've tested it. For now: it's buyable, the specs are published, and the tier you pick is a budget decision more than a capability guess.

Provider

Specs & key facts

What it isOpenAI's flagship LLM — next generation after GPT-5[src]
VariantsSix live: sol · sol-pro · terra · terra-pro · luna · luna-pro[src]
AvailabilityGenerally available since July 9, 2026 (preview announced June 26, 2026)[src]
AccessPurchasable via API — all six variants listed on OpenRouter[src]
Context window1,050,000 tokens[src]
Max output128,000 completion tokens[src]
Input modalityText + image + file → text[src]
LicenseClosed source / proprietary[src]

Capabilities

Reasoning / analysisYes (flagship positioning)
CodingYes
Multi-tier variantsSol / Terra / Luna, each with a -pro variant
Multimodal inputText, image, and file input; text output
Open weightsNo (closed source)
API accessYes — generally available since July 9, 2026

How to use it

  1. 1GPT-5.6 is generally available — pick a tier by budget: Luna ($1/$6 per 1M tokens) for high-volume work, Terra ($2.50/$15) as the default, Sol ($5/$30) when you want the top of the line.
  2. 2Call it by model id — openai/gpt-5.6-sol, gpt-5.6-terra, or gpt-5.6-luna (each also has a -pro variant listed at the same price) — via the OpenAI API or through OpenRouter.
  3. 3The 1,050,000-token context window means you can put whole repos or document sets in the prompt; output is capped at 128,000 tokens, so chunk long generations.
  4. 4Input accepts text, images, and files, so you can skip a separate vision or document-parsing step for many pipelines.
  5. 5Benchmark your own workload across the three tiers before committing — the 5x price spread between Luna and Sol is the whole point of the tier split.

Pricing

Sol / Sol Pro

$5.00 in · $30.00 outper 1M tokens

Top capability tier. The gpt-5.6-sol-pro variant carries the same listed per-token price as gpt-5.6-sol.

Terra / Terra Pro

$2.50 in · $15.00 outper 1M tokens

Balanced middle tier — half the Sol price. gpt-5.6-terra-pro is listed at the same price as gpt-5.6-terra.

Luna / Luna Pro

$1.00 in · $6.00 outper 1M tokens

Cheapest tier — one fifth of Sol on input and output. gpt-5.6-luna-pro is listed at the same price as gpt-5.6-luna.

Six GPT-5.6 variants are live and purchasable: sol, sol-pro, terra, terra-pro, luna, luna-pro. Per 1M tokens (input / output): Sol $5.00 / $30.00 · Terra $2.50 / $15.00 · Luna $1.00 / $6.00 — the -pro variants carry the same listed price as their base tier. Prices read directly from the OpenRouter public models API, verified 2026-07-21.

Pros & cons

Pros

  • Three-tier naming (Sol/Terra/Luna) signals clearer capability-cost positioning than before.
  • Generally available and purchasable via API since July 9, 2026 — no waitlist.
  • 1,050,000-token context window with text, image, and file input across the whole line.
  • 5x price spread between Luna and Sol lets you match tier to workload instead of overpaying by default.
  • OpenAI flagship pedigree — if GPT-5 was any indicator, the ceiling here is high.

Cons

  • No public benchmark numbers from OpenAI comparing the tiers to each other or to GPT-5.
  • OpenAI hasn't published what actually differentiates the -pro variants — they carry the same listed price as their base tier.
  • Sol at $30 per 1M output tokens is expensive for high-volume generation.
  • Closed source means no weight access, no fine-tuning without OpenAI's tooling.

Alternatives

FAQ

Sources

  1. 1.Preview announcement date (June 26, 2026), three-tier naming (Sol / Terra / Luna), OpenAI flagship positioninghttps://openai.com/index/previewing-gpt-5-6-sol/Verified 2026-07-08
  2. 2.Three-tier structure and positioning (Sol flagship, Terra balanced, Luna fastest/cheapest)https://openai.com/index/previewing-gpt-5-6-sol/Verified 2026-07-08
  3. 3.General availability date (2026-07-09); six purchasable variants (gpt-5.6-sol, sol-pro, terra, terra-pro, luna, luna-pro); per-tier input/output pricing; 1,050,000-token context window; 128,000 max completion tokens; text + image + file input modalityhttps://openrouter.ai/api/v1/modelsVerified 2026-07-21

More coverage

News & first-looks about this release. Coming soon.
Head-to-head comparisons. Coming soon.