Chasing Next
Explore LessonsAI UpdatesBlogAbout
Log inFor TeamsStart for Free
Chasing Next

Learn AI by doing.

Start for FreeFor TeamsAboutFAQUpdatesBlog
Sign UpSign InSupport
2026 Moonswell Marketing LLC·Terms·Privacy··
All Updates

Model Updates

New and updated AI models from every major provider.

Last Week
GoogleGooglemodel
Link·May 19, 2026

Gemini 3.5 Pro

New top tier Gemini model, currently in testing and available broadly next month.

  • Best for: Deep reasoning and complex agentic workflows that 3.5 Flash can't handle reliably
  • When to use it: Your task needs frontier reasoning or agent-driven action
  • Who benefits: Developers and Gemini app users on all plans.
GoogleGooglemodel
Link·May 19, 2026

Gemini Omni

New multimodal video generation model, that can also create custom avatars from your own voice and likeness.

  • Best for: Creating videos or using avatars that look and sound like you
  • When to use it: You need to create or edit video content
  • Who benefits: Google AI Plus, Pro, and Ultra subscribers using the Gemini app
Microsoft CopilotMicrosoft Copilotmodel
Link·May 19, 2026

GPT-5.5 Models in Microsoft 365 Copilot Chat

GPT-5.5 Instant and GPT-5.5 Thinking are now selectable models inside Copilot Chat.

Microsoft Copilot
  • Best for: Picking the right reasoning depth per task inside Copilot Chat
  • When to use it: You want Thinking for complex analysis or Instant for quick everyday work
  • Who benefits: Microsoft 365 Copilot users who already prefer GPT-5.5 over default model
GoogleGooglemodel
Link·May 19, 2026

Gemini 3.5 Flash

New mid-tier Gemini model in the 3.x family, announced at Google I/O 2026.

Google
  • Best for: Fast multimodal generation tasks where latency matters more than reasoning depth
  • When to use it: You need quick interactive outputs in the Gemini app or via API
  • Who benefits: Developers and Gemini app users who want a faster default model
May 2026
GoogleGooglemodel
Link·May 12, 2026

Gemini 3.1 Flash-Lite (Generally Available)

Google's low-cost, low-latency Gemini model is now generally available, with response times under two seconds for full replies.

  • Best for: high-volume tasks where response speed and per-call cost dominate model choice.
  • When to use it: you're building agents, classifiers, or tool-callers that run at scale.
  • Who benefits: developers running real-time agents, customer support automations, or IDE assistants.
OpenAIOpenAImodel
Link·May 7, 2026

GPT-Realtime-2, Translate, and Whisper

Three new realtime audio models in the API for voice agents that reason, translate, and transcribe live.

  • Best for: Voice agents that need to reason, translate across languages, or transcribe speech live over the Realtime API.
  • When to use it: Existing voice products hitting limits of older Realtime models on tool use, recovery, or live multilingual translation.
  • Who benefits: Developers shipping voice agents for support, scheduling, dubbing, or transcription who need GPT-5-class reasoning with low latency.
OpenAIOpenAImodel
Link·May 5, 2026

GPT-5.5 Instant

New default ChatGPT model with fewer hallucinations, shorter responses.

OpenAI
  • Best for: everyday ChatGPT use where accuracy and concise answers matter
  • When to use it: fact-sensitive prompts in medicine, law, finance, or any task where ChatGPT used to ramble
  • Who benefits: all ChatGPT users
April 2026
PerplexityPerplexitymodel
Link·Apr 24, 2026

GPT-5.5 in Perplexity

GPT-5.5 is now available in Perplexity for Max subscribers and is the default orchestration model in Perplexity Computer.

  • Best for: Perplexity-native research and Computer agent runs that benefit from GPT-5.5's planning
  • When to use it: you want OpenAI's newest model inside Perplexity rather than going to ChatGPT
  • Who benefits: Perplexity Pro and Max subscribers
OpenAIOpenAImodel
Link·Apr 23, 2026

GPT-5.5

New OpenAI model built for long time-span computer tasks, rolling out in ChatGPT and Codex.

  • Best for: multi-step agent work that spans the browser, files, docs, and code
  • When to use it: you need a model that plans, uses tools, and checks its own work across long tasks
  • Who benefits: Plus, Pro, Business, and Enterprise ChatGPT users and paid Codex developers
PerplexityPerplexitymodel
Link·Apr 23, 2026

Kimi K2.6 in Perplexity

Moonshot AI's open-weight Kimi K2.6 model is now selectable for Perplexity Pro and Max subscribers.

  • Best for: Perplexity users who want an open-weight alternative to GPT and Claude for search and reasoning queries
  • When to use it: you want to compare Kimi against Perplexity's other model options
  • Who benefits: Pro and Max subscribers exploring different model behaviors inside Perplexity