The GPT-5.6 family (Sol, Terra, Luna), GPT-5.5/5.4 generations, and unified reasoning systems: 1M+ context on frontier tiers, 90% prompt caching discounts, and 50% batch discounts.
25 models tracked · cheapest input: Text Embedding 3 (Small) at $0.02/M · official pricing page
OpenAI's GPT-5.6 family spans flagship reasoning (Sol, Cyber) down to ultra-cheap high-volume tiers (Luna), with the broadest ecosystem integration, 90% prompt caching and a 50% batch API. It is the default when ecosystem maturity and model breadth matter more than the absolute lowest rate card.
gpt-5.6-sol
Input /M
$5.00
Output /M
$30.00
Context
1.1M
OpenAI's frontier flagship: deepest reasoning, 1.05M context window, 128k output, and Sol-tier throughput for complex professional tasks.
gpt-5.6-terra
Input /M
$2.00
Output /M
$12.00
Context
1.1M
The balanced 5.6 tier: near-Sol quality at $2/$12 with the full 1.05M context window — the recommended default choice for production.
gpt-5.6-luna
Input /M
$0.20
Output /M
$1.20
Context
1.1M
The cheap 5.6 tier at $0.20/$1.20 — high-volume classification, routing, data extraction, and low-latency interactive agents.
gpt-5.6-cyber
Input /M
$12.50
Output /M
$75.00
Context
1.1M
Specialized cybersecurity reasoning model for vulnerability analysis, automated exploit synthesis prevention, and reverse engineering.
Input /M
$5.00
Output /M
$30.00
Context
512K
Established GPT-5.5 production generation for heavy enterprise workflows, structured data extraction, and autonomous agent loops.
gpt-5.5-pro
Input /M
$30.00
Output /M
$180.00
Context
512K
High-compute tier of GPT-5.5 delivering expanded compute budget per token for rigorous legal and financial analysis.
Input /M
$2.50
Output /M
$15.00
Context
256K
Solid general workhorse model offering balanced intelligence and high reliability at $2.50/$15.00 per 1M tokens.
gpt-5.4-mini
Input /M
$0.75
Output /M
$4.50
Context
256K
Compact GPT-5 tier optimized for subagents, automated tool orchestration, and high-frequency code completion.
gpt-5.4-nano
Input /M
$0.20
Output /M
$1.25
Context
128K
Ultra-low-cost utility model for instant document filtering, entity extraction, and sentiment scoring at $0.20/$1.25.
gpt-5.3-codex
Input /M
$1.75
Output /M
$14.00
Context
256K
Dedicated agentic coding model fine-tuned for repository-scale refactoring, unit test generation, and diff application.
Input /M
$20.00
Output /M
$80.00
Context
1M
Advanced multi-turn reasoning engine with extended chain-of-thought verification for breakthroughs in math and physics.
Input /M
$10.00
Output /M
$40.00
Context
1.1M
Frontier reasoning model generating extensive recursive verification steps for breakthrough scientific research.
o3-mini
Input /M
$1.10
Output /M
$4.40
Context
200K
High-speed reasoning model specialized in STEM, math, and competitive coding at $1.10/$4.40 with structured output.
o4-mini
Input /M
$1.10
Output /M
$4.40
Context
256K
Lightweight reasoning model with rapid thinking traces and multimodal image reasoning support.
Input /M
$2.50
Output /M
$10.00
Context
128K
OpenAI's previous-generation multimodal flagship. 128K context window with native vision and audio integration.
gpt-4o-mini
Input /M
$0.15
Output /M
$0.60
Context
128K
Cost-efficient small model at $0.15/$0.60 per 1M tokens. Standard benchmark model for high-throughput tasks.
Input /M
$15.00
Output /M
$60.00
Context
200K
First-generation deep reasoning flagship for complex mathematics and STEM problems.
o1-mini
Input /M
$1.10
Output /M
$4.40
Context
128K
Early compact reasoning model optimized for code generation and mathematical analysis.
gpt-4-turbo
Input /M
$10.00
Output /M
$30.00
Context
128K
Previous flagship GPT-4 generation with 128K context window and vision capabilities.
gpt-3.5-turbo-0125
Input /M
$0.50
Output /M
$1.50
Context
16.4K
Legacy workhorse model for simple chat and formatting tasks.
text-embedding-3-small
Input /M
$0.02
Output /M
$0.00
Context
8.2K
Highly efficient embedding model for vector search and RAG retrieval at $0.02 per 1M tokens.
text-embedding-3-large
Input /M
$0.13
Output /M
$0.00
Context
8.2K
OpenAI's most capable embedding model with up to 3072 dimensions for high-accuracy semantic search.
gpt-5.6-luna-pro
Input /M
$0.20
Output /M
$1.20
Context
1.1M
OpenAI reasoning model route with a 1.05M context window and higher-capacity Pro serving.
gpt-5.6-terra-pro
Input /M
$2.00
Output /M
$12.00
Context
1.1M
OpenAI reasoning model route with a 1.05M context window and higher-capacity Pro serving.
gpt-5.6-sol-pro
Input /M
$2.00
Output /M
$10.00
Context
1.1M
OpenAI reasoning model route with a 1.05M context window and higher-capacity Pro serving.
Other providers
All OpenAI prices verified Aug 28, 2026. Prices change frequently — each model page links to the authoritative source.