OpenAI shut down the Assistants API on August 26, 2026. Compare migration costs: Responses API, Azure OpenAI, Anthropic Claude, LangChain, open-source alternatives. Cost analysis at 3 usage tiers, free tier options, and migration timeline.
Published 2026-04-09 · Reviewed 2026-08-27, corrections outstanding · 6 migration paths compared · Figures compiled 2026-04-09 · 2 of 2 figures in the “OpenAI Pricing Changes Timeline” table come from our records for 1,493 developer tools · 15 OpenAI pricing changes tracked
What's happening: OpenAI sunset the Assistants API on August 26, 2026, and it is no longer available. Developers must migrate to the Responses API + Conversations API, or move to an alternative platform entirely. Azure is not a way out: Microsoft retired the Azure OpenAI Assistants API on the same date, August 26, 2026, and points agentic workloads at Microsoft Foundry Agent Service.
The cost question: Migration isn't just about code changes — it's about ongoing costs. Staying with OpenAI means the same token pricing but new API patterns. Switching providers can reduce or increase costs, depending on the models you move to. Open-source frameworks (LangChain, CrewAI) eliminate platform lock-in but require more engineering investment.
This guide covers: cost comparison at 3 usage tiers (hobby, production, scale), migration effort estimates, free tier options, and a recommended timeline — compiled by hand from vendor pricing pages.
What does each migration path cost at three usage levels? Hobby (occasional use, <1K requests/day), Production (steady workload, 1K–10K requests/day), and Scale (high-volume, 10K+ requests/day).
| Migration Path | Hobby /mo | Production /mo | Scale /mo | Migration Effort |
|---|---|---|---|---|
| OpenAI Responses API | $5–15 | $50–150 | $500–2,000 | Medium |
| Azure OpenAI Service | $10–25 | $100–300 | $1,000–5,000 | Low–Medium |
| Anthropic Claude API | $5–20 | $50–200 | $500–3,000 | Medium–High |
| Google Gemini API | $0 | $10–50 | $200–1,000 | Medium–High |
| LangChain + BYO LLM | $0 (local) – $10 | $20–100 | $200–1,000 | High |
| CrewAI | $0 (local) – $10 | $20–100 | $200–1,000 | High |
Each path trades off differently on cost, migration effort, and long-term flexibility.
Direct successor to Assistants API. Prompts replace Assistants, Conversations API replaces Threads. Token-based pricing unchanged. No automated migration tool — manual code rewrite required.
Free option: No GPT model is priced free · Best for: Existing OpenAI users who want minimal code changes
Same GPT models via Azure, but Microsoft retired the Azure OpenAI Assistants API on August 26, 2026 — the same date as OpenAI's. Azure workloads move to Microsoft Foundry Agent Service for agents, or the Azure OpenAI Responses API for inference only. Enterprise SLAs, data residency, and private endpoints included.
Free option: $200 Azure credit, 30 days · Best for: Enterprise users, teams already on Azure, compliance requirements
Drop-in replacement for many use cases. Current models: Claude Fable 5.1, Opus 5.5, Sonnet 5 and Haiku 4.5. Tool use (function calling) built in. 200K–1M context window vs GPT-4o's 128K. Prompt caching reduces costs 90% for repeated context.
Free option: Small free credits for new users · Best for: Teams wanting longer context, better reasoning, or tool use capabilities
The free tier covers the Gemini 3.x Flash and Flash-Lite models. Google publishes no free-tier limits; AI Studio shows each project's. Function calling, code execution, grounding with Google Search. OpenAI-compatible endpoint available for easier migration.
Free option: Free on Gemini 3.x Flash models · Best for: Cost-sensitive projects, teams wanting generous free tier
Open-source agent framework. Bring any LLM (OpenAI, Anthropic, local models via Ollama). LangGraph for stateful agents replaces Assistants' thread management. More control but more code to write.
Free option: Framework is free (MIT) · Best for: Teams wanting model flexibility, custom agent architectures
Open-source multi-agent framework. Define agents with roles and goals. BYO LLM (any OpenAI-compatible API). Good for complex workflows that used multiple Assistants coordinating. Hosted CrewAI Enterprise plan available.
Free option: Framework is free (MIT) · Best for: Multi-agent workflows, teams building specialized agent teams
If you're looking to reduce costs or run a prototype for free, these options have meaningful free tiers.
| Provider | Free Tier | Key Limit | Agent/Assistant Support |
|---|---|---|---|
| Google Gemini API | Free on 3.x Flash models | Limits shown in AI Studio | Function calling, code execution |
| OpenAI (Responses API) | No GPT model priced free | omni-moderation-latest is the only Free model | Prompts + tool use |
| Anthropic Claude | Small free credits for new users | Amount not stated | Tool use, 200K–1M context |
| LangChain | Fully free (MIT) | LLM API costs only | LangGraph agents, tool calling |
| CrewAI | Fully free (MIT) | LLM API costs only | Multi-agent orchestration |
| Azure OpenAI | $200 credit, 30 days | General Azure trial credit, not an agent entitlement | Microsoft Foundry Agent Service — Assistants API retired August 26, 2026 |
Key dates and recommended migration schedule.
| Date | Event | Urgency |
|---|---|---|
| August 2025 | OpenAI announces Assistants API deprecation | LOW |
| April 2026 | 4 months to shutdown — recommended migration start | MEDIUM |
| June 2026 | 2 months to shutdown — final migration window | HIGH |
| August 26, 2026 | Assistants API fully shut down — all calls return errors | HIGH |
OpenAI Responses API — same provider, same models, just new API patterns. Minimal code changes.
Google Gemini API — a free tier on its 3.x Flash models. OpenAI-compatible endpoint eases migration.
Azure OpenAI — same GPT models, enterprise SLAs, data residency. Rebuild on Microsoft Foundry Agent Service: Microsoft retired the Azure Assistants API on August 26, 2026 too, so this path is a change of platform, not a way to keep the old API.
Anthropic Claude — Claude Fable 5.1 for demanding reasoning and long-horizon agentic work. Tool use built in. Prompt caching reduces costs 90% for repeated context.
LangChain + BYO LLM — swap models anytime, no platform lock-in. More work upfront but full control.
CrewAI — if your Assistants setup used multiple coordinating agents, CrewAI's role-based agent model is a natural fit.
Recent pricing and policy changes we've tracked for OpenAI. See full change timeline for all tracked changes across all vendors.
| Date | Change | Impact |
|---|---|---|
| effective Sep 28, 2026 | gpt-3.5-turbo-instruct, babbage-002, davinci-002 and gpt-3.5-turbo-1106 shut down in the OpenAI API on September 28, 2026. OpenAI names gpt-5.6-terra as the replacement. Source ↗ | LOW |
| effective Sep 24, 2026 | Sora 2 video generation in the OpenAI API shut down on September 24, 2026: OpenAI removed the Videos API and the sora-2 and sora-2-pro models, and names no replacement. Source ↗ | MEDIUM |
| effective Aug 26, 2026 | Assistants API deprecated, full shutdown August 26, 2026. Developers must migrate to Responses API + Conversations API Source ↗ | HIGH |
| effective May 12, 2026 | dall-e-2 and dall-e-3 were removed from the OpenAI API on 2026-05-12. OpenAI names gpt-image-2, gpt-image-1 or gpt-image-1-mini as substitutes, and gpt-image-1 itself shuts down on 2026-10-23. Source ↗ | HIGH |
| effective May 12, 2026 | The Realtime API beta was removed on 2026-05-12; OpenAI's generally available Realtime API replaces it. Source ↗ | HIGH |
| effective Mar 20, 2024 | Between 2024-03-13 and 2024-03-20, OpenAI stopped giving new API accounts the $5 free trial credit, which could be used during an account's first 3 months. New accounts now have to buy prepaid credits, $5 minimum, to use any paid model. Source ↗ | HIGH |
These are announced changes that have not taken effect yet, listed soonest first.
| Date | Change | Impact |
|---|---|---|
| effective Oct 23, 2026 | The OpenAI API keeps running. The gpt-3.5-turbo, gpt-4, gpt-4-turbo, gpt-4-1106-preview, gpt-4.1-nano, gpt-4o-2024-05-13, gpt-image-1, o1, o1-pro, o3-mini and o4-mini models, with their fine-tuned versions, can be called until October 23, 2026, when OpenAI shuts down their access. OpenAI names a substitute for each model: gpt-3.5-turbo to gpt-5.6-terra; gpt-4, gpt-4-turbo, gpt-4-1106-preview and gpt-4o-2024-05-13 to gpt-5.6-sol; gpt-4.1-nano to gpt-5.6-luna; gpt-image-1 to gpt-image-2.5-sunburst or gpt-image-2.5-flare; o1 to gpt-5.6-sol; o1-pro to gpt-5.6-sol with reasoning.mode: pro; o3-mini to gpt-5.6-sol; o4-mini to gpt-5.6-terra. Source ↗ | MEDIUM |
| effective Nov 30, 2026 | The OpenAI API keeps running. Agent Builder, the Evals platform and the v1/prompts API can be used until November 30, 2026. Existing evals become read-only on October 31, 2026. ChatKit remains available. Migrate from Agent Builder to the Agents SDK or ChatGPT Workspace Agents. Migrate from Evals to Promptfoo. Move reusable prompt content into application code. Source ↗ | LOW |
| effective Dec 1, 2026 | The OpenAI API keeps running. The gpt-image-1-mini, gpt-image-1.5 and chatgpt-image-latest models can be called until December 1, 2026. OpenAI's substitutes are gpt-image-2.5-sunburst or gpt-image-2.5-flare. Source ↗ | LOW |
| effective Dec 11, 2026 | The OpenAI API keeps running. gpt-5-2025-08-07 is available until December 11, 2026, replaced by gpt-5.6-sol. gpt-5-mini-2025-08-07 is available until December 11, 2026, replaced by gpt-5.6-terra. gpt-5-nano-2025-08-07 is available until December 11, 2026, replaced by gpt-5.6-luna. gpt-5-pro-2025-10-06 is available until December 11, 2026, replaced by gpt-5.6-sol with reasoning.mode: pro. o3-2025-04-16 is available until December 11, 2026, replaced by gpt-5.6-sol. o3-pro-2025-06-10 is available until December 11, 2026, replaced by gpt-5.6-sol with reasoning.mode: pro. Source ↗ | MEDIUM |
| effective Jan 6, 2027 | The OpenAI API keeps running. The tts-1, tts-1-hd, gpt-4o-mini-tts-2025-03-20 and gpt-4o-mini-tts-2025-12-15 models can be called until January 6, 2027. OpenAI's replacement for each: gpt-realtime-2.1-mini. Source ↗ | LOW |
| effective Jan 14, 2027 | The OpenAI API keeps running. The gpt-4o (gpt-4o-2024-08-06 and gpt-4o-2024-11-20), gpt-4o-mini, gpt-4.1 and gpt-4.1-mini models, with their fine-tuned versions, can be called until January 14, 2027, when OpenAI shuts down their access. OpenAI names gpt-6-luna as the replacement for all four. Source ↗ | MEDIUM |
| effective Jan 20, 2027 | The OpenAI API keeps running. The gpt-realtime, gpt-4o-realtime, gpt-realtime-mini, gpt-4o-mini-realtime, gpt-audio, gpt-4o-audio, gpt-audio-mini, gpt-4o-mini-audio and gpt-4o-mini-transcribe-2025-03-20 models can be called until January 20, 2027. OpenAI's substitutes: gpt-realtime-2.1 for gpt-realtime and gpt-4o-realtime; gpt-realtime-2.1-mini for gpt-realtime-mini and gpt-4o-mini-realtime; gpt-audio-1.5 for gpt-audio, gpt-4o-audio, gpt-audio-mini and gpt-4o-mini-audio; gpt-4o-mini-transcribe-2025-12-15 for gpt-4o-mini-transcribe-2025-03-20. Source ↗ | LOW |
| effective Feb 26, 2027 | The OpenAI API keeps running. The whisper-1, gpt-4o-transcribe, gpt-4o-mini-transcribe and gpt-4o-transcribe-diarize models can be called until February 26, 2027. OpenAI's substitute for each is gpt-live-transcribe or gpt-transcribe. Source ↗ | MEDIUM |
| effective Apr 1, 2027 | The OpenAI API keeps running. The gpt-5.3-codex, gpt-5.1 and gpt-5.4-nano models can be called until April 1, 2027. OpenAI's replacements: gpt-6-sol for gpt-5.3-codex and gpt-5.1; gpt-6-luna for gpt-5.4-nano. Source ↗ | MEDIUM |