Black Forest Labs' image generation and editing model with bounding-box layouts, multiple reference images and 4K output
GPT-6 Sol
New· 4OpenAI's middle-tier GPT-6 model, with a 1,050,000-token context and text and image input, served as gpt-6-sol
About GPT-6 Sol
OpenAI announced GPT-6 Sol on 22 September 2026, together with GPT-6 Luna, and describes it as the middle tier of the GPT-6 family. In the API it is called gpt-6-sol. The headline from OpenAI is a price cut: the model is sold at half the rate of GPT-5.6 Sol, a cut that OpenAI credits to caching and inference improvements, not to a smaller model.
Capabilities listed in the documentation
The documentation page gives a context window of 1,050,000 tokens and up to 128,000 output tokens. The model takes text and images and returns text, and its knowledge cutoff is 20 April 2026. The same announcement raised the default cache hit rate and set the discount on cached input reads at 90 percent.
Where it was available at launch
At launch, Plus, Pro, Business, Enterprise and Edu users could reach it in ChatGPT Work and Codex, and developers could call it through the API. It was not in Chat on the first day.
Independent measurements
On the Artificial Analysis Intelligence Index (version 4.3.2), measured independently, it scores 48. Output speed was recorded at 126.0 tokens per second, and finishing the index took 77 million output tokens, with a blended rate worked out on a 7:2:1 cache-hit, input and output mix.
OpenAI's own figures
OpenAI reports 33.2 percent on AutomationBench 1.0.6 at xhigh effort, 56.4 percent on Agents' Last Exam V1 at max effort, 68.8 percent on DeepSWE v1.1 and 60.5 percent on OSWorld 2.0 offline at xhigh. These come from OpenAI's own evaluation. It also prints a cost per task beside the score, and puts GPT-6 Sol at a fraction of the cost per task of a rival model on AutomationBench. That comparison is worth reproducing on your own workload before relying on it.
Lifecycle
OpenAI publishes a deprecations page with dated shutdowns. When this entry was written, the page showed the Videos API being removed on 24 September 2026 and twelve legacy snapshots on 23 October 2026.
TrustList summary
GPT-6 Sol sits in the middle of OpenAI's GPT-6 range. OpenAI announced it on 22 September 2026 alongside GPT-6 Luna, and serves it through the API under the identifier gpt-6-sol.
What GPT-6 Sol offers
- Text and image input: accepts text and images and returns text.
- Long context: a 1,050,000-token window and up to 128,000 output tokens.
- ChatGPT products: available in ChatGPT Work and Codex to Plus, Pro, Business, Enterprise and Edu subscribers, but not in Chat on launch day.
- Independent score: Artificial Analysis rates it 48 on version 4.3.2 of its Intelligence Index, at 126.0 output tokens per second.
Clients
API developers, and ChatGPT Work and Codex users on the plans listed above.
Key facts
- Knowledge cut-off: 20 April 2026.
- Caching: the default cache hit rate was raised in the same announcement.
- Lifecycle: OpenAI's deprecations page lists the Videos API for removal on 24 September 2026.
- Benchmarks: OpenAI's own figures are its own evaluation and are not independently checked.
Summary written by TrustList from the business's own published information, Nov 2026.
Benchmarks & AI stats
| Benchmark | Official | Community avg |
|---|---|---|
| Artificial Analysis Intelligence Index v4.3.2llm | 48 index | - |
“Official” values are editor-approved and feed the ranking. “Community avg” is the mean of member submissions (shown for transparency; it never affects the ranking until an editor approves a value).
Sign in to add a benchmark score for this model.
GPT-6 Sol launches
Trust Score
An earned signal from verification, reviews, awards, transparency and engagement - the vendor can't buy it.
Updated 9/23/2026
Request a demo or quote from GPT-6 Sol
Reviews
Write the first review of GPT-6 Sol
Used it? Your experience helps other buyers decide.
Write a reviewQuestions & answers
No questions yet. Be the first to ask about GPT-6 Sol.
GPT-6 Sol alternatives
Similar listings buyers compare against this one.
Microsoft AI's real-time speech-to-text model for voice agents, transcribing 60 languages with continuous language detection
Open-weight translation model from Bilibili's Index team, covering 150 languages and following glossary and style instructions
Cantina Security's first open-weight security model, a post-trained GLM-5.3-Flash for vulnerability research and exploit checks
Google's open multimodal embedding model that maps text, images, audio and video into one vector space, under Apache 2.0
Perplexity's open-weight decision model that picks among given options for routing and checks in AI systems
Open-source 2B decision model from AWS's Strands Agents project that picks options and returns scores with a reliability figure
Anthropic's smallest and fastest Claude 5.5 model, available on its own API and three cloud platforms
More on TrustList
Everything here links back to the same verified catalogue. Pick your next stop.
- Write a reviewHelp other buyers decide
- CompaniesAgencies, consultancies and IT service providers, ranked by verified reviews.
- ProductsSoftware and SaaS with pricing, features, integrations and alternatives.
- ArticlesGuides, comparisons and research from the editorial desk and the community.
- AwardsAnnual recognition decided by verified reviews and an independent jury.
- LaunchesNew products and releases, voted up by the community every day.
- RequestsBuyers describe what they need; vendors respond directly.
- PeopleReviewers, authors and makers with public profiles.
- ComparePut up to four listings side by side before you shortlist.