Instruction Tuning (FLAN)

Deployment timeline

1. GPT-4o

   Deployed 2026-02-16 · Velocity 4y

   “GPT-4o follows instruction-tuned models like InstructGPT, enabling strong zero-shot task following.”

   high
2. GPT-5

   Deployed 2026-02-16 · Velocity 4y

   “GPT-5 follows the instruction-following paradigm established by its predecessors, which were instruction-tuned.”

   high
3. GPT-5.2 Pro

   Deployed 2026-02-17 · Velocity 4y

   “As a flagship OpenAI model, it follows the established instruction-following paradigm refined from InstructGPT and GPT-4.”

   high
4. Claude 3

   Deployed 2026-02-18 · Velocity 4y

   “Claude 3 is instruction-tuned to follow diverse instructions, a core capability mentioned in documentation.”

   high
5. Gemini 3 Pro

   Deployed 2026-02-19 · Velocity 4y

   “Gemini models are instruction-tuned, building on the FLAN line of work.”

   high
6. Gemini 3.1

   Deployed 2026-02-20 · Velocity 4y

   “Gemini models are instruction-tuned, building on Google's FLAN tradition.”

   medium
7. Claude 3.5 Sonnet

   Deployed 2026-02-23 · Velocity 4y

   “Claude models are instruction-tuned on diverse tasks to follow instructions and generalize to new tasks.”

   high
8. Claude Sonnet 4.6

   Deployed 2026-02-25 · Velocity 4y

   “Claude models are instruction-tuned to follow user instructions, a standard practice for Anthropic.”

   medium
9. GPT-5.3

   Deployed 2026-02-26 · Velocity 4y

   “GPT-5.3 follows OpenAI's instruction-tuned lineage (InstructGPT) for strong zero-shot generalization.”

   medium
10. Claude 4.5

    Deployed 2026-02-26 · Velocity 4y

    “Claude models are instruction-tuned to follow natural language instructions effectively.”

    high
11. Gemini 3 Flash

    Deployed 2026-02-27 · Velocity 4y

    “Gemini models are instruction-tuned, building upon the FLAN instruction-tuning methodology developed by Google.”

    high
12. Kimi K2.5

    Deployed 2026-03-04 · Velocity 5y

    “Kimi models are instruction-tuned for conversational ability, aligning with FLAN-style training.”

    medium
13. Gemini 3.1 Flash-Lite

    Deployed 2026-03-05 · Velocity 5y

    “Gemini models are instruction-tuned, building on the FLAN instruction-tuning methodology.”

    high
14. Claude 3.5 Opus

    Deployed 2026-03-18 · Velocity 5y

    “Claude models are instruction-tuned; Claude 3.5 Opus shows strong instruction following.”

    high
15. GLM-5.1

    Deployed 2026-03-21 · Velocity 5y

    “GLM-5.1 is instruction-tuned on diverse tasks following FLAN methodology.”

    medium
16. Qwen 3.6

    Deployed 2026-03-31 · Velocity 5y

    “Qwen models are instruction-tuned on a large collection of datasets.”

    high
17. GPT-5.4-Cyber

    Deployed 2026-04-16 · Velocity 5y

    “GPT models are instruction-tuned to follow diverse tasks, aligning with the FLAN paradigm.”

    high