GPT-6 Sol API Pricing Calculator & Benchmarks

GPT-6 Sol is OpenAI’s balanced frontier model for complex coding, reasoning, agentic workflows, computer use, and demanding professional applications.

GPT-6 Sol, released 22 September 2026, is a multimodal model from OpenAI. It pairs a 1,050,000-token context window with up to 128,000 tokens of output, taking text, image and file input and returning text. Training data ends 20 April 2026, and the model reasons step by step. It costs between $1.00 and $8.00 per million input tokens. GPT-6 Sol is the successor to GPT-5.1 and GPT-5.3 Codex.

Last updated Oct 27, 2026

Never miss an AI pricing or launch update

Get notified about new model launches, pricing changes, provider updates, deprecations, and cost-saving insights.

πŸ”’ We respect your privacy. Unsubscribe anytime.

Model Details

Released
Sep 22, 2026
Knowledge Cutoff
Apr 20, 2026
Context Length
1,050,000
Max Output
128,000
Modalities
Text Image File → Text
Capabilities
Tool use Structured outputs Prompt caching Code execution Computer use

Where is GPT-6 Sol a perfect fit?

GPT-6 Sol combines strong reasoning with lower inference costs, supporting coding, computer use, tool calling, long-context analysis, and autonomous agent workflows at production scale. The model is suitable for below use-cases:
Complex software engineering β€” implementation, debugging, refactoring, and large codebases
- Coding agents β€” autonomous multi-step development workflows
- Computer-use agents β€” interacting with websites and graphical applications
- Tool-heavy workflows β€” function calling, web search, file search, and computer use
- Long-context analysis β€” large documents, repositories, and research material
- Demanding reasoning β€” technical, analytical, and professional tasks
- Production agentic systems β€” applications requiring a balance of capability, latency, and cost

Quick Model Estimate

(USD 2.0000 per 1M tokens)
(USD 10.0000 per 1M tokens)

Your GPT-6 Sol Cost Estimate

πŸ’° Total Cost

β€”

for 1000 input + 1000 output tokens

πŸ“₯ Input (1000 Γ— $2.000000) β€”
πŸ“€ Output (1000 Γ— $10.000000) β€”

Cost Breakdown

πŸ“₯ Input πŸ“€ Output

Prices are watched for changes and checked against the provider's own page.

Pricing

Provider ↕
Modality ↕
Service Tier ↕
Input Price
(per 1M tokens)
↕
Output Price
(per 1M tokens)
↕
Cached Input
(per 1M tokens)
↕
Cached Output
(per 1M tokens)
↕
Context Size ↕
View
OpenAI LogoOpenAITextStandard$2.0000$10.0000$0.2000$2.5000Up to 272,000 tokens→
OpenAI LogoOpenAITextStandard$4.0000$15.0000$0.4000$5.0000272,000+ tokens→
OpenAI LogoOpenAITextBatch$1.0000$5.0000$0.1000$1.2500Up to 272,000 tokens→
OpenAI LogoOpenAITextBatch$2.0000$7.5000$0.2000$2.5000272,000+ tokens→
OpenAI LogoOpenAITextFlex$1.0000$5.0000$0.1000$1.2500Up to 272,000 tokens→
OpenAI LogoOpenAITextFlex$2.0000$7.5000$0.2000$2.5000272,000+ tokens→
OpenAI LogoOpenAITextPriority$4.0000$20.0000$0.4000$5.0000Up to 272,000 tokens→
OpenAI LogoOpenAITextPriority$8.0000$30.0000$0.8000$10.0000272,000+ tokens→
Perplexity LogoPerplexityTextStandard$2.0000$10.0000$0.2000N/AUp to 272,000 tokens→
Perplexity LogoPerplexityTextStandard$4.0000$15.0000$0.4000N/A272,000+ tokens→

Benchmarks

Scores from standardized evaluations by Artificial Analysis and Design Arena. Higher is better β€” the indices summarize overall ability, while the detailed scores break down performance on individual benchmarks.

Artificial Analysis

Higher is better Β· benchmarked by Artificial Analysis
Index scores by variant
Detailed scores
47.6
Intelligence Index
Overall intelligence across reasoning, knowledge & math evals
Better than 77% of 32 models
β€”
Coding Index
Coding ability across software-engineering evals
β€”
Agentic Index
Tool use & multi-step agent task performance
Reasoning
Humanity's Last Exam 47.9%

Expert-level questions across many academic domains

CritPt 30.9%

Unpublished physics research reasoning problems

AA-LCR 83.7%

Long-context reasoning across large inputs

Coding
SciCode 57.6%

Research-level scientific coding tasks

Knowledge
GDPval 50.4%

Economically valuable, real-world knowledge work

Omniscience Accuracy 54.5%

Breadth of factual knowledge across domains

Omniscience Non-Hallucination 39.9%

How reliably the model avoids fabricated answers

All variants
Variant Humanity's Last Exam AA-LCR SciCode CritPt GDPval Omniscience Accuracy Omniscience Non-Hallucination
GPT-6 Sol (Max) 47.9% 83.7% 57.6% 30.9% 50.4% 54.5% 39.9%
GPT-6 Sol (Xhigh) 46.3% 81.3% 55.1% 28% 47.8% 53.8% 41.1%
GPT-6 Sol (High) 44.1% 83.7% 54.9% 25.4% 44.8% 53.7% 41.9%
GPT-6 Sol (Medium) 41% 82.3% 53.8% 24.6% 42.3% 53.4% 43.2%
GPT-6 Sol (Low) 34.9% 79.3% 50.2% 16.3% 34.8% 51.2% 49.3%
GPT-6 Sol (Non-reasoning) 18.4% 64% 47.3% 4% 37.6% 45.2% 16%

FAQs about GPT-6 Sol

How much does GPT-6 Sol cost per 1M tokens?

GPT-6 Sol costs between $1.00 and $8.00 per million input tokens, and between $5.00 and $30.00 per million output tokens. The range covers Standard, Batch, Flex and Priority pricing.

What does a typical workload cost with GPT-6 Sol?

1,000 requests of 2,000 input and 500 output tokens each β€” 2,000,000 input and 500,000 output tokens in total β€” costs $4.50 with GPT-6 Sol at its lowest rates. Use the calculator on this page for your own volumes.

Which provider is cheapest for GPT-6 Sol?

OpenAI has the lowest input rate for GPT-6 Sol, at $1.00 per million tokens. Across all providers: OpenAI from $1.00 and Perplexity from $2.00 per million input tokens.

Is there cheaper GPT-6 Sol pricing than the standard rate?

Yes. Batch pricing for GPT-6 Sol starts at $1.00 per million input tokens against $2.00 on the standard tier, about 50% lower.

What is GPT-6 Sol's context window?

GPT-6 Sol has a context window of 1,050,000 tokens, and returns up to 128,000 tokens in a single response.

What input types does GPT-6 Sol support?

GPT-6 Sol accepts text, image and file input, and returns text.

What is GPT-6 Sol's knowledge cutoff?

GPT-6 Sol's training data runs to 20 April 2026. It has no built-in knowledge of events after that date.