Ask HN: Is it just me or is Claude Code getting worst?

8 points · 8 visible top comments · 2026-04-28 16:21:56 UTC

news.ycombinator.com · Read Story HN original

Is it just me or is Claude Code getting worst and worst, since they introduced the 1 million context on the 4.6, things start to go real bad, am I the only one ? PS: I am still paying the 200 euros monthly max

Comments

sovenyr · 2026-04-28 16:31:53 UTC

can't say it become worse, but at some point it stops be so useful as it was before - it looks like magic disapear

itdar · 2026-04-28 16:37:52 UTC

I can't say which is better, but 4.6 was the most intense

omer_30300 · 2026-04-28 16:42:12 UTC

Yeah buddy, Claude Code is honestly getting worse lately. It's been giving me buggy or incorrect code for my projects as well

runjake · 2026-04-28 17:12:44 UTC

It's not just you. It's getting much worse. There is a lot of talk on X about it, along with hypotheses and evidence-based testing.

Kai_Build · 2026-04-28 17:20:29 UTC

It is probably not just you, but I would be careful about blaming the 1M context window specifically.

Anthropic’s recent postmortem described several Claude Code regressions around default reasoning effort, context/thinking retention, and a system prompt change to reduce verbosity. Those seem more likely to explain the “less careful / forgetful / worse follow-through” behavior than the context window alone.

I would compare the same task in a fresh session, with the effort setting fixed, and ideally against a few repeatable tasks from your own codebase. Otherwise it is very hard to tell whether the regression is the model, Claude Code’s harness, context management, or just a stale session.

alegd · 2026-04-28 17:21:32 UTC

not just you. I use it daily and the quality has been inconsistent lately. Some sessions are great, others it feels like it forgot how to read my codebase. For 200 euros/month you'd expect to be nothing but awesome.

lgl · 2026-04-28 17:30:10 UTC

Due to the copilot nerfing recently I've switched to codex and gpt 5.4 (and now especially 5.5) have been doing pretty great.

But even codex has these super weird time limits. It's really starting to show that these companies must have been losing a ton of money with all the recent limits and degration.

I'm still on the "camp" that most of these unicorns will be F'ed by open and local models in the next few years, at least in these coding/chatbox niches and then they'll just be perpetually (re)searching for AGI :shrug:

idempotent_ · 2026-04-28 18:19:03 UTC

Claude the model is still insanely great IF (and perhaps, ONLY IF) you are willing to fork over the money for the API and use a harness like OpenCode.

Claude Code itself is complete trash. They had a massive headstart and now are routinely lapped by open source harnesses and then they STILL double down on not allowing e.g. OpenCode usage with the Max plan. Meanwhile, OpenAI lets you use whatever harness you want and its a beast. I recently did some testing and OpenAI's Pro plan on an opencode harness (GPT 5.5 XHigh) with parallel agent delegation absolutely smokes Claude Code 4.7 Max. These days Claude Code can barely even remember its CLAUDE.MD instructions. I'd say Opus 4.7 Max API is slightly better than GPT 5.5 XHigh, but not nearly enough that the API token price is at all justified.

Claude, I think is still better for business things like document generation, design, etc. especially via claude.ai interface (GDrive integrations and things like that are very useful). But for code generation and dev workflows, Claude Code is dropping the ball so hard its starting to look like a generational fumble.

ApolloRising · 2026-04-28 20:42:32 UTC

Why do you feel the OpenCode harness is better than the Claude code one? Just curious what you feel it is doing better?

idempotent_ · 2026-04-28 22:28:36 UTC

OpenCode harness is so good I'm surprised one of the big players hasn't bought them outright. Essentially their harness:

* Removes all the system prompt cruft and bullshit that CC pumps into the prompt and pollutes context and shit like "adaptive thinking"

* Is extremely good at keeping the model aligned with AGENTS.MD and opencode.json and using all the features available there (parallel agents, sub-sub agents, etc)

For example, I'm working on a repo with 5 distinct components and I have a specialized agent for each component. CLAUDE.MD is just a markdown file where I say "Hey Claude always use X agent for X component. X agent has this prompt blah blah" and then pray Claude remembers to use it. opencode.json is a structured file used by the harness and it has ALWAYS coerced the model to use it, including being able for the agents to delegate subagents in parallel etc.

This makes a massive difference. So if I have a feature that touches multiple components, OpenCode rips through it with the specialized subagents while Claude sits their spinning its wheels and occasionally remembering theres a specialized agent and then maybe once in a blue moon it will do it in parallel.

With CC I feel like I need to do all these invocations and coercions. OpenCode, once you've got your opencode.json and agents defined, just works.