# GPT-4.1

OpenAI, released 2025-04-14

**Verdict: recommended.** The default migration target for GPT-4 Turbo workloads: same request shape, lower cost, 1M-token context — convert legacy function_call payloads to the tools format and move on. Last verified 2026-06-12.

## Pricing

- Input: $2 / 1M tokens
- Output: $8 / 1M tokens
- Context: 1M tokens

## Strengths

- Drop-in replacement for GPT-4 Turbo chat and completion workloads at lower per-token cost
- Long-context processing up to 1M tokens
- Workloads pinned to the legacy function_call parameter — GPT-4.1 rejects it

## Benchmarks

- MMLU: 90.2% (https://openai.com/index/gpt-4-1/)
- SWE-bench Verified: 54.6% (https://openai.com/index/gpt-4-1/)

## Sources

- [OpenAI GPT-4.1 announcement](https://openai.com/index/gpt-4-1/)
- [OpenAI Pricing](https://openai.com/api/pricing/)

Canonical: https://neomanex.com/models/gpt-4.1