ChatGPT Model Catalog 2025: Every GPT-5.1, GPT-4o, GPT-4.1, and GPT-3.5 Variant with Plan Mapping Explained
- Graziano Stefanelli
- 5 hours ago
- 3 min read
ChatGPT now spans a diverse lineup of models—ranging from ultra-fast GPT-5.1 Instant to multimodal GPT-4o and long-context GPT-4.1 mini—each tied to specific subscription tiers, context windows, and specialty tools.
Understanding which model appears in which plan unlocks reliable performance, cost control, and feature access across everyday chat, vision tasks, deep reasoning, and enterprise automation.
··········
··········
GPT-5.1 variants provide speed, depth, and agent tools for modern ChatGPT sessions.
The GPT-5.1 family includes Instant for rapid replies, Thinking for deeper reasoning, and Pro for the highest-effort mode with advanced agent features.
Instant is the default picker for free, Plus, and Business plans, offering a 128 K-token context window and 16 K output tokens.
Thinking and Pro expand reasoning depth to a 196 K-token window, but appear only in Plus, Pro, and Enterprise tiers.
·····GPT-5.1 Model Lineup
Variant | Context Window | Key Trait |
Instant | 128 K tokens | Fast responses |
Thinking | 196 K tokens | Deeper logic |
Pro | 196 K tokens | Agent tools & research |
··········
··········
The GPT-4 family remains the backbone for multimodal and vision-heavy tasks.
GPT-4o and GPT-4o-mini power real-time text-vision-audio features, with 128 K windows and near-instant streaming for voice mode.
GPT-4.1 adds a 1 M-token context in API workflows, while the ChatGPT UI caps effective context at 256 K for stability.
Turbo and legacy GPT-4 models continue in the API but are hidden in the consumer picker pending deprecation.
·····GPT-4 Family Overview
Model | Core Feature | Context Window |
GPT-4o | Real-time multimodal | 128 K tokens |
GPT-4o-mini | Low-latency, free fallback | 128 K tokens |
GPT-4.1 | Long-context API | 1 M tokens API / 256 K UI |
GPT-4 Turbo Legacy | Cost-optimized | 128 K tokens |
··········
··········
GPT-4.1 mini extends enterprise context while lowering cost.
Introduced in mid-2025, GPT-4.1 mini offers a 1 M-token window in the API and 256 K in ChatGPT, giving businesses a large-context alternative to premium GPT-5 plans at a reduced price.
It slots in as the default fallback for Plus and Business when GPT-5.1 thinking quotas are exhausted, preserving context continuity for long documents.
··········
··········
Legacy GPT-3.5 models persist as low-cost engines for high-volume automation and free-tier backups.
GPT-3.5-turbo-0125 and 3.5-light-0613 remain in the API and Free plan legacy tab, servicing inexpensive workloads and acting as the final fallback once GPT-5.1 quotas expire.
They provide 16 K–32 K token windows, suitable for lightweight chatbots, logging, or batch transformations where premium logic is unnecessary.
··········
··········
Special-purpose engines add code, image, and video generation beyond core chat.
Code Interpreter and Advanced Codex agents run on GPT-5.1 Pro back ends, enabling data analysis and file manipulation in Pro and Enterprise tiers.
DALL·E 4 powers image creation for Plus users and above, while Sora 1 video generation enters research preview for Pro subscribers.
Voice Mode relies on GPT-4o stream inference, ensuring fast text-to-speech and speech-to-text performance across all plans.
··········
··········
Plan mapping aligns each ChatGPT subscription tier to its default and selectable models.
The Free plan defaults to GPT-4o-mini, then offers up to ten GPT-5.1 Instant messages every five hours before falling back to GPT-3.5.
Plus users start on GPT-4o-mini but can choose GPT-5.1 Instant, GPT-5.1 Thinking, GPT-4o, or GPT-4.1 as needed.
Pro becomes the only tier with GPT-5.1 Pro by default, while Business and Enterprise unlock the full picker plus elevated quotas, memory extensions, and agent-mode toggles.
·····Subscription-to-Model Matrix
Plan | Default Model | Optional Upgrades |
Free | GPT-4o-mini | GPT-5.1 Instant (quota), GPT-3.5 |
Plus | GPT-4o-mini | GPT-5.1 Instant / Thinking, GPT-4o, GPT-4.1 |
Pro | GPT-5.1 Instant | GPT-5.1 Pro, GPT-4.1 |
Business / Enterprise | GPT-5.1 Instant | GPT-5.1 Thinking / Pro, GPT-4o, GPT-4.1 |
··········
··········
Selecting the right model balances context needs, modality, cost, and speed.
Daily chat and lightweight vision tasks run smoothly on GPT-4o-mini or GPT-5.1 Instant.
Long-document work, enterprise compliance, and research benefit from GPT-4.1 mini’s 1 M-token API window or GPT-5.1 Thinking’s deeper reasoning.
Agent builders and advanced analysts move to GPT-5.1 Pro for code tools, larger memory, and multi-agent orchestration.
Meanwhile, legacy GPT-3.5 remains the economical choice for mass automation where frontier intelligence is unnecessary.
··········
FOLLOW US FOR MORE
··········
··········
DATA STUDIOS
··········