OpenAI shut down the Assistants API on August 26, 2026. Compare migration paths: Responses API, Claude, Gemini, open-source frameworks. Free tier comparison for 10+ AI API providers with stability ratings.
Published 2026-04-02 · Not yet reviewed · Figures in the tables below come from our records for 1,498 developer tools · 19 OpenAI pricing changes tracked
What’s happening: OpenAI deprecated the Assistants API in favor of the Responses API. OpenAI shut down the Assistants API on August 26, 2026; it is no longer available. All developers using Threads, persistent Assistants, Code Interpreter, or File Search through the Assistants API must migrate.
Key insight: The Responses API is a direct replacement with feature parity plus new capabilities (MCP support, deep research, computer use). Most developers should migrate to Responses API first — then evaluate whether to diversify to other providers for cost or capability reasons.
Our data says: OpenAI’s stability rating is volatile based on 19 tracked pricing changes — including free tier credit removal, ChatGPT ad insertion, and now this API sunset. Developers building on OpenAI should have a diversification plan.
The Assistants API provided a stateful, session-based interface for building AI agents. OpenAI is replacing it with the Responses API — a stateless, more flexible approach. Here’s what breaks and what replaces it:
| Assistants API Feature | What Breaks | Responses API Replacement |
|---|---|---|
| Threads | Persistent conversation state deleted | Conversations API (or manage context yourself) |
| Persistent Assistants | Assistant objects no longer exist | Prompts + system instructions per request |
| Code Interpreter | Built-in code execution removed | Code interpreter tool in Responses API |
| File Search | Vector store file search removed | File search tool in Responses API |
| Function Calling | Same concept, different API shape | Function calling (compatible pattern) |
| Run Lifecycle | Polling-based run management removed | Streaming-first, simpler lifecycle |
Four paths depending on your constraints. The first path (stay with OpenAI) requires the least code changes. The others offer different trade-offs between cost, capability, and vendor independence.
Direct replacement with feature parity. Threads → Conversations API. Assistants → prompts + system instructions. Code Interpreter and File Search tools carry over. New capabilities: MCP support, deep research, computer use.
Best for: Teams heavily invested in OpenAI ecosystem, production apps where minimizing migration risk matters most
Claude (best reasoning, long context), Gemini (free tier, multimodal), Cohere (RAG/enterprise search). Each has native tool use/function calling. Requires API integration changes but not architectural rewrites.
Best for: Teams wanting to reduce single-vendor dependency, those needing specific capabilities (long context, multimodal, search)
LangChain, CrewAI, AutoGen, or custom orchestration with open-source models via Ollama, vLLM, or llama.cpp. Full control over the stack. No vendor lock-in. Run locally or on your own infrastructure.
Best for: Teams with ML ops capability, privacy-sensitive workloads, cost-conscious at scale
OpenRouter or LiteLLM as a unified API layer across multiple LLM providers. Switch models without code changes. Compare pricing and latency across providers. Reduces lock-in to any single API.
Best for: Teams wanting flexibility to switch providers, cost optimization across multiple models
All 12 alternative AI API providers compared. The tier and paid-rate columns are read from each provider's record in our index at request time, and the stability ratings come from our stability dashboard. A dash means the record carries no paid rate; the link goes to the vendor's own pricing page. Click provider names for full vendor profiles.
| Provider | Recorded Tier | Paid Rate | Tool Use | Code Execution | Stability |
|---|---|---|---|---|---|
| OpenAI (Responses API) | Pay-as-you-go | gpt-6-luna $0.10/$0.50 – gpt-6-astra $10.00/$50.00 per MTok | Native (function calling) | Code interpreter tool | volatile |
| |||||
| Anthropic Claude API | Pay-as-you-go | Claude Haiku 5.5 $0.10/$0.50 – Claude Fable 5.1 $10/$50 per MTok | Native (tool use) | Computer use, code execution | volatile |
| |||||
| Google Gemini API | Free (Reduced) | Gemini 3.1 Flash-Lite $0.25/$1.50 – Gemini 3.1 Pro Preview $2/$12 per MTok | Native (function calling) | Code execution tool | watch |
| |||||
| GitHub Models | Retired | — | Via hosted models | No built-in | retired |
| OpenRouter | Free | — vendor pricing | Model-dependent | Model-dependent | stable |
| |||||
| Cohere | Free | Command R7B $0.0375/$0.15, Command R $0.15/$0.60 per MTok | Native (tool use) | No built-in | stable |
| |||||
| Groq | Free | gpt-oss-20b $0.075/$0.30, gpt-oss-120b $0.15/$0.60, Qwen3.8 27B $0.80/$4.00 per MTok | Native (function calling) | No built-in | volatile |
| |||||
| Fireworks AI | Free Credits | $0.05/$0.20 – Fireworks serverless serves Kimi K3 $3.00/$15.00 per MTok | Native (function calling) | No built-in | unrated |
| Together AI | Pay-as-you-go | gpt-oss-120B $0.15/$0.60 – Together AI serverless serves Kimi K3 $3.00/$15.00 per MTok | Native (function calling) | No built-in | volatile |
| |||||
| Mistral AI | Pay-as-you-go | Ministral 3 $0.1/$0.1 – Mistral Medium 3.5 $1.5/$7.5 per MTok | Native (function calling) | No built-in | volatile |
| |||||
| DeepSeek API | Pay-as-you-go | DeepSeek-V4.1-Flash $0.30/$1.20, DeepSeek-V4-Pro $1.32/$3.96 per MTok | Native (function calling) | No built-in | volatile |
| |||||
| Cerebras | Trial | GPT OSS 120B $0.35/$0.75, Qwen 3.8 27B $0.99/$1.49 per MTok | Via Llama models | No built-in | volatile |
| |||||
OpenAI’s stability rating is volatile. We’ve tracked 19 changes. Here’s the full history:
| 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 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 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-2.6-sunburst or gpt-image-2.6-flare. Source ↗ | LOW |
| effective Dec 1, 2026 | OpenAI's promotional pricing for gpt-5.6-sol on the OpenAI API ends on November 30, 2026. Until then, gpt-5.6-sol costs $4.00 per million input tokens, $0.40 per million cached input tokens and $20.00 per million output tokens on Standard processing, and $2.00, $0.20 and $10.00 on Flex and Batch. OpenAI has not said what gpt-5.6-sol costs from December 1, 2026. OpenAI's note covers gpt-5.6-sol on the OpenAI API only and says nothing about Azure OpenAI. Source ↗ | MEDIUM |
| effective Dec 11, 2026 | The OpenAI API keeps running. gpt-5 and gpt-5-2025-08-07 are available until December 11, 2026, replaced by gpt-6.1-sol. gpt-5-mini and gpt-5-mini-2025-08-07 are available until December 11, 2026, replaced by gpt-6.1-terra. gpt-5-nano and gpt-5-nano-2025-08-07 are available until December 11, 2026, replaced by gpt-6-luna. gpt-5-pro and gpt-5-pro-2025-10-06 are available until December 11, 2026, replaced by gpt-6.1-sol with reasoning.mode: pro in the Responses API. o3 and o3-2025-04-16 are available until December 11, 2026, replaced by gpt-6.1-sol. o3-pro and o3-pro-2025-06-10 are available until December 11, 2026, replaced by gpt-6.1-sol with reasoning.mode: pro in the Responses API. o3-mini is a different model from o3, and OpenAI shut it down on October 23. 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 6, 2027 | The OpenAI API keeps running. From January 6, 2027, no organization can create new fine-tuning jobs on OpenAI's self-serve fine-tuning platform. Since May 7, 2026, organizations that never ran fine-tuning cannot create jobs. Since July 2, 2026, organizations that have not run inference on a fine-tuned model in the past 60 days cannot either. Fine-tuned models keep serving inference until their base model is deprecated. Source ↗ | MEDIUM |
| 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 1, 2027 | The OpenAI API keeps running. The gpt-5.5-cyber-preview model can be called until February 1, 2027, when OpenAI removes it from the API. OpenAI tells users to move to the most capable cyber model available to them. Source ↗ | LOW |
| effective Feb 10, 2027 | The OpenAI API keeps running. The gpt-image-2 model and its gpt-image-2-2026-04-21 snapshot can be called until February 10, 2027, when OpenAI removes them from the API. OpenAI names gpt-image-2.6-sunburst or gpt-image-2.6-flare as the replacement. Source ↗ | MEDIUM |
| 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. The /v1/audio/translations endpoint also shuts down on February 26, 2027. OpenAI's replacement is /v1/realtime/translations. 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 |
OpenAI Responses API — direct migration, same vendor, feature parity plus new capabilities. Start here.
Anthropic Claude API — Claude Fable 5.1 $10/$50 per MTok at the top of the lineup our record holds. Best for complex multi-step agent workflows that need strong reasoning.
Google Gemini API — free tier with rate limits, multimodal input (images, audio, video). Best for prototyping and low-volume production.
GitHub Models is recorded as Retired — GitHub ended it, so it is no longer a way to reach several models for free. OpenRouter is the multi-model route still open in our index, recorded as Free.
Cohere — Command R7B $0.0375/$0.15 per MTok, the lowest paid rate our index carries for any provider on this page. That is 3× cheaper than gpt-6-luna at the same volume.
Groq or Cerebras — ultra-fast inference on open-source models. Best for latency-sensitive applications.
OpenRouter — unified API across 500+ models from 80+ providers. Switch models without code changes. Compare pricing in real-time.
Cohere — purpose-built for retrieval-augmented generation with native document parsing. Best for knowledge-heavy agent workflows.
How we track this data: AgentDeals monitors free tier changes across 1,498 developer tools in 60 categories. Stability ratings are computed from our deal changes database — OpenAI is classified as volatile based on 19 tracked changes including free tier removal, limit reductions, and API deprecation.
Vendor data verification: Free tier details were read from vendor pricing pages on 2026-04-02. Last full verification: March 2026. Provider capabilities (tool use, code execution) were read from official API documentation on 2026-04-02.
For real-time data, use our stability dashboard, Atom feed, or MCP server. Full dataset available via REST API.