OpenAI models in Big-AGI
GPT on your own key: reasoning effort up to max, verbosity and Pro mode per request, plus hosted search, code interpreter and image tools.
All supported OpenAI models
ModelContextInputOutputReleased
GPT-6.1 TerraNEWFast visual reasoning, computer use, and writing at half the price of GPT-6.1 Sol. Reasoning can be turned off. 1M token context.1.1M$1.00$5.00Oct 2026GPT-6.1 SolNEWNear-Astra coding, computer use, and professional work at a fifth of Astra's price. Reasoning always on. 1M token context.1.1M$2.00$10.00Sep 2026GPT-6 LunaNEWLowest-cost GPT-6 tier, for focused, high-volume tasks. Succeeds GPT-5.6 Luna at half the price. 1M token context.1.1M$0.10$0.50Sep 2026GPT-6 SolNEWMiddle GPT-6 tier, below Astra: complex coding and agentic workflows. Succeeds GPT-5.6 Sol at half the price, with about half its factual errors. Reasoning can…1.1M$2.00$10.00Sep 2026GPT-6 AstraMost capable OpenAI model, built for the hardest end-to-end work: reasoning, coding, computer use, research, and document creation. Fewer output tokens per tas…1.1M$10.00$50.00Sep 2026GPT-5.6 TerraBalanced model for efficient, high-volume everyday work. Competitive with GPT-5.5 at a fraction of the cost. 1M token context.1.1M$2.00$12.00Jul 2026GPT-5.6 SolFlagship GPT-5.6 tier, for agentic coding, science, and cybersecurity. 1M token context.1.1M$4.00$20.00Jul 2026GPT-5.6 LunaFastest, most affordable GPT-5.6 model for high-volume work. Strong capability at the lowest cost in the family. 1M token context.1.1M$0.20$1.20Jul 2026ChatGPT Instant400K$5.00$30.00May 2026GPT-5.5New baseline for complex production workflows. Stronger task execution, more precise tool use, more efficient reasoning with fewer tokens. 1M token context.1.1M$5.00$30.00Apr 2026GPT-5.5 ProMost capable model for complex tasks. Uses more compute for smarter, more precise responses on the hardest problems.1.1M$30.00$180.00Apr 2026GPT-5.4 MiniHOTStrongest mini model for coding, computer use, and subagents. GPT-5.4-class intelligence at lower cost and latency.400K$0.75$4.50Mar 2026GPT-5.4 NanoCheapest GPT-5.4-class model for simple high-volume tasks like classification and data extraction.400K$0.20$1.25Mar 2026GPT-5.4Most capable and efficient frontier model for professional work. Native computer use, improved reasoning, coding, and agentic workflows with 1M token context.1.1M$2.50$15.00Mar 2026GPT-5.4 ProMost capable model for complex tasks. Uses more compute for smarter, more precise responses on difficult problems.1.1M$30.00$180.00Mar 2026GPT-5.3 InstantdeprecatedGPT-5.3 Instant model, previously powering ChatGPT.128K$1.75$14.00Mar 2026GPT Audio 1.5Best voice model for audio in, audio out with Chat Completions. Accepts audio inputs and outputs.128K$2.50$10.00Feb 2026GPT-5.3 CodexdeprecatedMost capable agentic coding model. Combines frontier coding performance of GPT-5.2-Codex with reasoning and professional knowledge of GPT-5.2. ~25% faster.400K$1.75$14.00Feb 2026GPT-5.2 InstantdeprecatedGPT-5.2 Instant model, previously powering ChatGPT.128K$1.75$14.00Dec 2025GPT-5.2 CodexdeprecatedGPT-5.2 optimized for long-horizon, agentic coding tasks in Codex or similar environments. Supports low, medium, high, and xhigh reasoning effort settings.400K$1.75$14.00Dec 2025GPT-5.2Most capable model for professional work and long-running agents. Improvements in general intelligence, long-context, agentic tool-calling, and vision.400K$1.75$14.00Dec 2025GPT-5.2 ProSmartest and most trustworthy option for difficult questions. Uses more compute for harder thinking on complex domains like programming.400K$21.00$168.00Dec 2025GPT-5.1 Codex MaxdeprecatedOur most intelligent coding model optimized for long-horizon, agentic coding tasks.400K$1.25$10.00Nov 2025GPT-5.1HOTThe best model for coding and agentic tasks with configurable reasoning effort.400K$1.25$10.00Nov 2025GPT-5.1 CodexdeprecatedA version of GPT-5.1 optimized for agentic coding tasks in Codex or similar environments.400K$1.25$10.00Nov 2025GPT-5.1 Codex MinideprecatedSmaller, faster version of GPT-5.1 Codex for efficient coding tasks.400K$0.25$2.00Nov 2025GPT-5.1 InstantdeprecatedGPT-5.1 Instant with adaptive reasoning. More conversational with improved instruction following.128K$1.25$10.00Nov 2025GPT-5 Search APIUpdated web search model in Chat Completions API. 60% cheaper with domain filtering support.400K$1.25$10.00Oct 2025GPT-5 ProVersion of GPT-5 that uses more compute to produce smarter and more precise responses. Designed for tough problems.400K$15.00$120.00Oct 2025GPT Audio MinideprecatedCost-efficient audio model. Accepts audio inputs and outputs via Chat Completions REST API.128K$0.60$2.40Oct 2025GPT-5 CodexdeprecatedA version of GPT-5 optimized for agentic coding in Codex.400K$1.25$10.00Sep 2025GPT AudioFirst generally available audio model. Accepts audio inputs and outputs, and can be used in the Chat Completions REST API.128K$2.50$10.00Aug 2025GPT-5The best model for coding and agentic tasks across domains.400K$1.25$10.00Aug 2025GPT-5 ChatGPTdeprecatedGPT-5 model used in ChatGPT.128K$1.25$10.00Aug 2025GPT-5 MiniA faster, more cost-efficient version of GPT-5 for well-defined tasks.400K$0.25$2.00Aug 2025GPT-5 NanoFastest, most cost-efficient version of GPT-5 for summarization and classification tasks.400K$0.05$0.40Aug 2025o3 Deep ResearchdeprecatedOur most powerful deep research model for complex, multi-step research tasks.200K$10.00$40.00Jun 2025o4 Mini Deep ResearchdeprecatedFaster, more affordable deep research model for complex, multi-step research tasks.200K$2.00$8.00Jun 2025o3 ProVersion of o3 with more compute for better responses. Provides consistently better answers for complex tasks.200K$20.00$80.00Jun 2025o3A well-rounded and powerful model across domains. Sets a new standard for math, science, coding, and visual reasoning tasks.200K$2.00$8.00Apr 2025o4 MiniLatest o4-mini model. Optimized for fast, effective reasoning with exceptionally efficient performance in coding and visual tasks.200K$1.10$4.40Apr 2025GPT-4.1Flagship GPT model for complex tasks. Major improvements on coding, instruction following, and long context with 1M token context window.1M$2.00$8.00Apr 2025GPT-4.1 MiniBalanced for intelligence, speed, and cost. Matches or exceeds GPT-4o in intelligence while reducing latency by nearly half and cost by 83%.1M$0.40$1.60Apr 2025GPT-4.1 NanoFastest, most cost-effective GPT 4.1 model. Delivers exceptional performance with low latency, ideal for tasks like classification or autocompletion.1M$0.10$0.40Apr 2025o1 ProA version of o1 with more compute for better responses. Provides consistently better answers for complex tasks.200K$150.00$600.00Mar 2025GPT-4o Mini Search PreviewdeprecatedGPT-4o Mini model optimized for web search capabilities. Alias still serving after its snapshot's 2026-07-23 shutdown.128K$0.15$0.60Mar 2025GPT-4o Search PreviewdeprecatedGPT-4o model optimized for web search capabilities. Alias still serving after its snapshot's 2026-07-23 shutdown.128K$2.50$10.00Mar 2025o3 MiniLatest o3-mini model snapshot. High intelligence at the same cost and latency targets of o1-mini. Excels at science, math, and coding tasks.200K$1.10$4.40Jan 2025o1Previous full o-series reasoning model.200K$15.00$60.00Dec 2024GPT-4o miniAffordable model for fast, lightweight tasks. GPT-4o Mini is cheaper and more capable than GPT-3.5 Turbo.128K$0.15$0.60Jul 2024GPT-4oSnapshot that supports Structured Outputs. gpt-4o currently128K$2.50$10.00May 2024GPT-4 TurboGPT-4 Turbo with Vision model. Vision requests can now use JSON mode and function calling. gpt-4-turbo currently128K$10.00$30.00Apr 20243.5-TurbodeprecatedThe latest GPT-3.5 Turbo model with higher accuracy at responding in requested formats.16K$0.50$1.50Jan 20243.5-TurbodeprecatedGPT-3.5 Turbo model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more.16K$1.00$2.00Nov 2023GPT-4Snapshot of gpt-4 from June 13th 2023 with improved function calling support. Data up to Sep 2021.8K$30.00$60.00Jun 2023GPT-4deprecatedSnapshot of gpt-4 from June 13th 2023 with improved function calling support. Data up to Sep 2021.8K$30.00$60.00Jun 20233.5-TurboThe latest GPT-3.5 Turbo model with higher accuracy at responding in requested formats.16K$0.50$1.50May 2023GPT-6 AstradeprecatedMost capable OpenAI model, built for the hardest end-to-end work: reasoning, coding, computer use, research, and document creation. Fewer output tokens per tas…1.1M$10.00$50.00-? [gpt live transcribe]deprecatedUnknown, please let us know the ID. Assuming a context window of 128k tokens, and a maximum output of 4k tokens.128K---? [gpt transcribe]deprecatedUnknown, please let us know the ID. Assuming a context window of 128k tokens, and a maximum output of 4k tokens.128K---? [ra gpt 5.6 sol]deprecatedUnknown, please let us know the ID. Assuming a context window of 128k tokens, and a maximum output of 4k tokens.128K---61 models · sorted by release date · prices in USD per 1M tokens · refreshed hourlyCompare every model across vendors ->
Get started in 3 steps
1
Create an API key at the OpenAI console.
2
Paste it into Big-AGI's model settings.
3
Start chatting, or Beam it against other models and fuse the answers.
Running OpenAI in Big-AGI
Add your OpenAI API key and use OpenAI's models at OpenAI's own API rates. Big-AGI isn't a reseller: no markup and no intermediary, so the bill comes straight from OpenAI to your account.
- Your key, your billing. Usage is billed by OpenAI directly, nothing routes through us.
- Reasoning effort and verbosity. Set both per request - effort from none up to max - so you tune depth and length instead of accepting a default.
- Function calling and structured outputs. JSON-schema responses your code can parse without cleanup.
- Hosted tools, tuned not guessed. Web search takes an adjustable context size and geolocation, and image generation and editing runs a real GPT Image pipeline with mask-based edits.
Why not just ChatGPT?
ChatGPT gives you one model in one box. Big-AGI runs OpenAI's models next to Claude, Gemini, Grok, and anything else you've connected, in the same conversation, something ChatGPT has no reason to offer.
It also hands you the dials ChatGPT hides: reasoning effort and verbosity are real parameters on OpenAI's API, not settings we invented. Your keys stay in your browser, not inside an OpenAI account. Big-AGI itself needs no account either, paste a key and start.
OpenAI bills per token too, so a light user can spend less in a month than a flat subscription, and a heavy user never hits an artificial cap, just a bill for tokens used. Beam parallel runs cost more tokens than a single ChatGPT reply, the honest tradeoff for comparing frontier models side by side.
Your keys and your data
Turn on Direct Connection under OpenAI's Advanced settings and the browser calls OpenAI directly, skipping the Big-AGI server, whenever your key lives client-side and OpenAI's API allows it. Your keys stay in your browser. Chats are stored locally first and sync only if you turn it on. The AI Inspector shows the exact request before it goes out: model, parameters, token counts, and a cost estimate.
OpenAI in Beam
Beam runs OpenAI's models against Claude, Gemini, or any other model you've connected, in parallel, then hands you Fusions: several strategies that combine, cross-check, and synthesize the parallel answers, which beats just picking the best one. Parallel runs use more tokens than a single chat.
Bring your OpenAI key. Keep control.
Your key, your data, your choice of model. Big-AGI's Open branch is open source and self-hostable, so you can check exactly how OpenAI is called.
Launch Big-AGIBIG-AGI
Resources
© 2026 Token Fabrics·Built with passion in San Diego