Pulse Changelog

Tracking every iteration of intelligence. From quantization tweaks to massive architectural shifts.

Performance

+30%

Nodes Added

15

Uptime

99.99%

v2.0.4

Stable

Key Shifts

  • Optimized Neural Orchestrator for 30% faster sharding
  • Enhanced INT4 Quantization precision for Transformer models
  • Implemented Vercel-native sharding endpoints
  • Resolved high-latency edge cases in EU clusters

v2.0.0

Major

Key Shifts

  • Full rewrite of the Python core to modular TypeScript
  • Integrated Razorpay for global credit distribution
  • Launched the Advanced Neural Analytics dashboard
  • Added support for Apple Neural Engine directly via CoreML export

v1.8.2

Patch

Key Shifts

  • Memory leak fix in persistent sharding buffers
  • Improved error handling for malformed ONNX artifacts
  • Standardized API response contracts for better SDK sync