Skip to content
All systems operational0 AI providers monitored, polled every 2 minutes
Live status

GPT-5.6 Sol

Flagship

by OpenAI

GPT-5.6 Sol is the flagship in OpenAI's GPT-5.6 family, previewed on June 26, 2026 and generally available since July 9, 2026. Access at launch was intentionally narrow: at the request of the US Government, only roughly 20 pre-approved organizations were granted entry to the limited preview. On August 21, 2026 OpenAI cut the Sol rate for the first time since GPT-5.5 set $5/$30 back in April. Input fell from $5 to $4 per million tokens, output from $30 to $20, and cached input from $0.50 to $0.40. OpenAI lists the new rate as promotional and good through at least November 21, 2026, so budget for a revert rather than treating it as the new floor. The cut covers the pay-as-you-go API, Codex credits, and eligible ChatGPT Work plans; Pro, Plus, and Business subscription prices are unchanged. The context window is 1 million tokens. OpenAI positioned Sol for frontier reasoning and agentic work and disclosed partial benchmarks at preview (Sol Ultra at 91.9 percent on Terminal-Bench 2.1, base Sol at 88.8 percent), with a fast mode available at twice the standard rate for roughly 2.5 times the throughput. On September 22, 2026 OpenAI released GPT-6 Sol at $2/$10, half the promotional GPT-5.6 Sol rate and permanent rather than promotional.

Input Price

$4.00

per 1M tokens

Output Price

$20.00

per 1M tokens

Context Window

1M

tokens

Released

2026-07

API access

Capabilities

textvisiontool-usecodereasoning

Key Strengths

  • ✓1M token context window
  • ✓Promotional $4/$20 per 1M tokens through November 21, 2026, down from $5/$30
  • ✓Cached input at $0.40 per 1M
  • ✓Frontier-class agentic and reasoning posture
  • ✓Sister to lower-cost Terra and Luna variants in the same family

Best For

  • ▸Long-horizon agentic workflows
  • ▸Frontier reasoning and research tasks
  • ▸Hardest-tier coding agents
  • ▸Limited-preview evaluation against GPT-5.5 and Claude Fable 5

Benchmark Scores

BenchmarkScoreDescription
SWE-bench96.2Real-world software engineering tasks from GitHub issues (SWE-bench Verified)
MMLU-Pro89.1General knowledge and reasoning across 57 subjects
GPQA Diamond94.6Graduate-level science questions verified by domain experts
OSWorld 2.062.6Computer use across real desktop applications and multi-step GUI tasks
BrowseComp90.4Agentic web search and browsing over hard-to-find facts
FrontierCode v1.147.5Agentic coding on frontier software engineering tasks (Main split)
Terminal-Bench 4.027.8Long-horizon agentic work in a terminal across software, science, ML, operations, hardware, security, and media (66 tasks, all-or-nothing verifiers)

Scores sourced from public benchmark datasets. See full benchmark leaderboard for all models.

Pricing Details

Input tokens

$4.00

per 1M tokens

Output tokens

$20.00

per 1M tokens

Estimated cost per 1K requests

$14.00

~1K input + ~500 output tokens avg

Prices are subject to change. Check the official documentation for current pricing. See the cost calculator for detailed estimates.

Related Models

View DocumentationCompare ModelsCost CalculatorFull Pricing Guide