Major Release
v2.5.0 — AIP-4 Launch
We're excited to announce the launch of AIP-4, our most capable model yet with breakthrough improvements in reasoning, coding, and multimodal understanding.
New Features
- AIP-4 model with 200K context window
- Vision capabilities for image understanding
- Function calling with parallel execution
- JSON mode for structured outputs
- System prompts for better control
Improvements
- 40% faster inference times
- Improved code generation accuracy
- Better multilingual support (100+ languages)
- Enhanced safety guardrails