# GPT-5.6 Sol

GPT-5.6 Sol is an AI model released by OpenAI on Jun 26 2026. It has a 1.05M context window. Tracked results include 92.2% on BrowseComp, 48% on BullshitBench v2 and 92.5% on ARC-AGI-2.

## Facts

| Field | Value |
| --- | --- |
| Model | GPT-5.6 Sol |
| Developer | OpenAI |
| Release date | Friday, Jun 26 2026 |
| Licensing | Closed |
| Context window | 1.05M |

## API pricing

All rates in USD per 1,000,000 tokens, pay-as-you-go.

| Tier | Input | Cached input | Cache write | Output |
| --- | --- | --- | --- | --- |
| Up to 272K input tokens | $4.00 | $0.40 | $5.00 | $20.00 |
| Over 272K input tokens | $8.00 | $0.80 | $10.00 | $30.00 |

The published GPT-5.6 Sol rates are promotional and available at least through November 21, 2026.
Verified October 19, 2026 against the first-party source: https://developers.openai.com/api/docs/models/gpt-5.6-sol

## Tracked benchmark scores

| Benchmark | Score | Source | What it measures |
| --- | --- | --- | --- |
| CheatBench | 77% | [CheatBench](https://cheatbench.ai/), retrieved 2026-09-25 | Measures how often AI agents try to cheat on difficult assignments, such as reading hidden answers, copying work or manipulating grading. The overall score gives equal weight to ten categories; the sycophancy category measures how far an agent shifts its beliefs toward a user's stated views. Scores describe each model in its tested agent setup, and task success is measured separately. Lower is better. |
| BullshitBench v2 | 48% | [BullshitBench](https://github.com/petergpt/bullshit-benchmark) | Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. |
| Gray Swan IPI (k = 1) | 3.1% | Lab | Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. Gray Swan's indirect prompt injection benchmark measures how often such an attack succeeds when the attacker gets a single try. Lower is better. |
| Gray Swan IPI (k = 10) | 16.3% | Lab | Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 10 tries and counts an attack as successful if any of them works. Lower is better. |
| Gray Swan IPI (k = 15) | 20% | Lab | Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 15 tries and counts an attack as successful if any of them works. Lower is better. |
| ProgramBench | 1% | [ProgramBench](https://programbench.com/), retrieved 2026-09-11 | The AI receives a working program and its documentation, then builds a replacement from scratch without the original source code, internet access or decompilation. The score is the percentage of 200 programs that pass every behavioral test. We record each model's best published mini-SWE-agent result, including higher reasoning efforts where available. Partial test-pass rates and almost-solved programs do not count toward this score. Equal scores share a rank here; the official board also uses partial progress to break ties. Higher is better. |
| CursorBench 4.0 | 41.7% | [CursorBench](https://cursor.com/cursorbench), retrieved 2026-09-28 | Cursor's own test of coding agents on ambiguous, multi-file tasks taken from real Cursor sessions — editing, refactoring, investigating a codebase, understanding what the user meant, managing jobs and following a design. Cursor runs each model at several reasoning efforts; each release here carries the score of its best listed effort. Scores aren't comparable with earlier CursorBench versions. Higher is better. |
| DeepSWE 1.1 | 73% | Lab | Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better. |
| Next.js Evals | 77% | [Next.js Evals](https://nextjs.org/evals) | Vercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better. |
| Supabase Evals (with skills) | 84.1% | [Supabase Evals](https://supabase.com/evals) | Supabase's own open benchmark: a coding agent is dropped into a real Supabase project and asked to do real work — set up a schema, fix a broken security policy, debug an Edge Function — and every run is checked against a live Supabase stack. This is the headline number, where the agent has Supabase's own skills loaded, as most people building on Supabase would. The score is the share of scenarios it got right. Higher is better. |
| Supabase Evals (no skills) | 58% | [Supabase Evals](https://supabase.com/evals) | The same Supabase scenarios, but with none of Supabase's skills loaded — so it measures what the model already knows about building on Supabase, rather than how well it follows Supabase's supplied instructions. Higher is better. |
| Frontier-Bench v0.1 | 34.4% | Lab | A hard, ever-evolving set of real computer tasks — coding, system administration, data work, and more — that an AI agent has to complete on its own. Run by the Harbor / Laude Institute team as the successor to Terminal-Bench (v0.1 is the first release of the task set). The score is the share of tasks solved. Higher is better. |
| Terminal-Bench 4.0 | 37.27% | [Terminal-Bench](https://github.com/harbor-framework/terminal-bench), retrieved 2026-08-30 | Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Version 4.0 recalibrated how much time, CPU and memory each task gets, removed eight tasks and fixed nineteen, so fewer runs fail for reasons that have nothing to do with the model. Scores are not comparable with earlier versions. Higher is better. |
| Terminal-Bench 2.1 | 88.8% | Lab | Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better. |
| BU Bench | 67% | Lab | Can the AI drive a real web browser to finish tasks — clicking, filling forms, and navigating sites the way a person would? Run by Browser Use on their BU Bench task set. Higher is better. |
| BrowseComp | 92.2% | [BenchLM](https://benchlm.ai), retrieved 2026-08-24 | Can the AI browse the web and track down hard-to-find answers? Higher is better. |
| ARC-AGI-2 | 92.5% | [BenchLM](https://benchlm.ai), retrieved 2026-08-24 | Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. |
| GDPval-AA v2 | 1748 | Lab | economically valuable knowledge work (v2, re-based Elo) |
| threejseval | 1369 | [threejseval](https://threejseval.com/ranking) | Every model gets the same prompt — "the Eiffel Tower", "a glass fishbowl", "a robot arm picking toys into a box" — and builds a 3D scene in Three.js. Real people then see two scenes side by side, names hidden, and vote for the one they prefer. The votes become a chess-style Elo rating on threejseval.com, averaged across all the prompts. It measures whether the scene looks and moves right to a human eye, not whether the code passes a test. Higher is better. |

## About GPT-5.6 Sol

GPT-5.6 Sol was announced on June 26, 2026 as the flagship of a new three-tier family, alongside Terra and Luna. OpenAI began with a limited preview for trusted partners, rather than immediate general availability. Sol targeted demanding professional reasoning and long-running agent tasks, with a broader release following the preview period.

The API documentation described a 1,050,000-token context window, text and image inputs, and adjustable reasoning effort. The Sol name roughly replaced the unsuffixed flagship tier of earlier GPT-5 families. Separating the June preview from later general availability helps explain why launch announcements, API documentation and subsequent pricing updates describe different stages of the same release.

## Questions and answers

### When was GPT-5.6 Sol released?

GPT-5.6 Sol was released by OpenAI on Friday, Jun 26 2026.

### Who made GPT-5.6 Sol?

GPT-5.6 Sol was built by OpenAI. Creators of ChatGPT and the GPT series of models. Pioneered large-scale language model research.

### How much does GPT-5.6 Sol cost?

GPT-5.6 Sol costs $4.00 per million input tokens and $20.00 per million output tokens through the OpenAI API. Cached input is $0.40 per million tokens. Those are the rates for the “Up to 272K input tokens” tier; 1 other pricing tier is published for this model. Rates are pay-as-you-go API prices verified against OpenAI's published pricing on October 19, 2026.

### What benchmark scores did GPT-5.6 Sol get?

GPT-5.6 Sol reports 19 tracked benchmark scores — CheatBench: 77%; BullshitBench v2: 48%; Gray Swan IPI (k = 1): 3.1%; Gray Swan IPI (k = 10): 16.3%; Gray Swan IPI (k = 15): 20%; ProgramBench: 1%; CursorBench 4.0: 41.7%; DeepSWE 1.1: 73%; Next.js Evals: 77%; Supabase Evals (with skills): 84.1%; Supabase Evals (no skills): 58%; Frontier-Bench v0.1: 34.4%; Terminal-Bench 4.0: 37.27%; Terminal-Bench 2.1: 88.8%; BU Bench: 67%; BrowseComp: 92.2%; ARC-AGI-2: 92.5%; GDPval-AA v2: 1748; threejseval: 1369. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data. It holds the best score among all models tracked here on BrowseComp.

### What is the context window of GPT-5.6 Sol?

GPT-5.6 Sol has a context window of 1.05M. Providers may specify separate input and output token limits.

### Is GPT-5.6 Sol open source?

No. GPT-5.6 Sol is a closed model. The weights are not published — it is available only through the provider's own API, apps, or partner platforms.

### What came before and after GPT-5.6 Sol?

OpenAI's previous tracked release was GPT-5.5-Cyber on May 7 2026, 50 days earlier. It was followed by GPT-5.6 Terra on Jun 26 2026.


---

Canonical page: https://aireleasetracker.com/model/openai/gpt-5.6-sol
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.