Skip to contentOceanToken

2 min read

29 Models Retired on November 10, 2026, and What to Use Instead

Their makers stopped serving 29 models, including o1, o3-mini, o4-mini, GPT-4 Turbo, GPT-3.5 Turbo and MiniMax-M1. Here is the model to switch to for each one.

By OceanToken Team

Short answer: On November 10, 2026 we removed 29 models from the OceanToken catalog because their makers no longer serve them. If your code calls one of them, change the model field to the replacement in the table below. Nothing else in your request needs to change, and no other model's price has changed.

Why these models were removed

We did not choose to retire these models. Their makers have stopped serving them, so no request to them can be answered any more, by us or anyone else. Rather than leave them in the catalog and return errors, we have taken them out.

What happens to a request for a retired model

A request that names one of these models now gets an error instead of a reply, and a request that gets an error is not charged. Switch to the replacement and the request works as before.

The 29 retired models and their replacements

Retired model Use instead
openai/o1 openai/gpt-5.5
openai/o1-pro openai/gpt-5.4-pro
openai/o3-mini openai/gpt-5.4-mini
openai/o3-mini-high openai/gpt-5.4-mini with reasoning_effort: "high"
openai/o4-mini openai/gpt-5.4-mini
openai/o4-mini-high openai/gpt-5.4-mini with reasoning_effort: "high"
openai/gpt-4-turbo openai/gpt-4.1
openai/gpt-3.5-turbo openai/gpt-5.4-nano
minimax/minimax-01 minimax/minimax-m3
minimax/minimax-m1 minimax/minimax-m3
minimax/minimax-m2 minimax/minimax-m2.7
minimax/minimax-m2.1 minimax/minimax-m2.7
minimax/minimax-m2-her minimax/minimax-m2.7
nousresearch/hermes-3-llama-3.1-405b nousresearch/hermes-4-405b
nousresearch/hermes-3-llama-3.1-70b nousresearch/hermes-4-405b
meta-llama/llama-3.1-70b-instruct meta-llama/llama-3.3-70b-instruct
upstage/solar-pro-3 upstage/solar-pro4
ibm-granite/granite-4.0-h-micro ibm-granite/granite-4.2-8b
qwen/qwen3-next-80b-a3b-thinking qwen/qwen3.8-27b
nvidia/nemotron-3-super-120b-a12b:free (free) nvidia/nemotron-3.5-lightning:free (free), or the paid nvidia/nemotron-3-super-120b-a12b
nvidia/nemotron-3.5-content-safety:free (free) nvidia/nemotron-3.5-content-safety (paid)
poolside/laguna-xs-2.1 moonshotai/kimi-k2.7-code
poolside/laguna-xs-2.1:free (free) moonshotai/kimi-k2.7-code
poolside/laguna-s-2.1 moonshotai/kimi-k2.7-code
poolside/laguna-s-2.1:free (free) moonshotai/kimi-k2.7-code
inception/mercury-2 deepseek/deepseek-v4.1-flash
inception/mercury-2.5 deepseek/deepseek-v4.1-flash
arcee-ai/trinity-large-thinking deepseek/deepseek-v4.1-flash
x-ai/grok-imagine-image-quality x-ai/grok-imagine-image-2.0

How we picked each replacement

Where the same maker still offers a newer model of the same kind, that is the replacement. Where a maker has stopped serving its whole line (Poolside, Inception and Arcee), we picked a model of the same kind from our catalog: Kimi K2.7 Code for coding work, and DeepSeek V4.1 Flash where speed and a low price matter most. Every replacement is in the catalog today, and its price is on the Models page.

A replacement is a newer or similar model, not an identical one, so its answers, speed and price can differ. If quality matters for your use, try a few of your own prompts on it before you switch everything over.

Questions

If you are not sure which model to move to, or you rely on one of these models for something the table does not cover, write to support@oceantoken.ai and we will help you choose.

  • Notice
  • Model retirements
  1. 1 min read

    8 Models Retired on November 16, 2026, and What to Use Instead

    Seed 2.0 Code, Seed 1.6, Seed 1.6 Flash, Qwen 2.5 7B, Qwen3 VL 30B Thinking, Hunyuan A13B, Step 3.5 Flash and LFM 2.5 2.6B (free) have left the catalog. Here is the model to switch to for each one.

    • Notice
    • Model retirements