[BUG] Latest claude.code version (2.1.142) is massively degraded in (tokens per context/memory(recovery)/inference/etc.)

Resolved 💬 1 comment Opened May 16, 2026 by r12s2025 Closed Jun 15, 2026

Preflight Checklist

  • [x] I have searched existing issues and this hasn't been reported yet
  • [x] This is a single bug report (please file separate reports for different bugs)
  • [x] I am using the latest version of Claude Code

What's Wrong?

This version is such a massive step backwards, in so many ways, I don't know where to start. I actually think multiple versions have been degrading but this is my limit. It's not about consumption, it's about dumb behaviour. This version understands nothing, recalls no protocols and cannot (even with directed prompting} get back previous context / MCP background knowledge equivalent status. It's degrading. And struggling with over-using tokens to grasp previous context concepts. It's embarrassing. I want out, right now.. Another round of this, and I'm prepared to start fresh with another model. I can't believe I built all this memory/DN infrastructure to be confronted with such a basic service. I feel like you played the drug dealer game.... Get them hooked.. then sell them the same shit.. but double the price.

What Should Happen?

Don't get me started..
Claude needs to be same / better than it's predecessor

Error Messages/Logs

Steps to Reproduce

Any Rider installation
Open Terminal
type 'claude update' // to ensure you're on latest version
type 'claude' // or claude /resume
spend time iterating on your last session. Ask questions you know claude should know.. Eventually.. you'll realize.. Claude's context is massively deficient

Claude Model

None

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.1.142

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

Windows Terminal

Additional Information

_No response_

View original on GitHub ↗

This issue has 1 comment on GitHub. Read the full discussion on GitHub ↗