chrismustcode

I’d be stunned if a 270m model could code with any proficiency.

If you have an iPhone with the semi-annoying autocomplete that’s a 34m transformer.

Can’t imagine a model (even if it’s a good team behind it) to do coding with 8x the parameters of a next 3/4 word autocomplete.

That was previous iterations, 2.5 is 1 million context window

https://ai.google.dev/gemini-api/docs/models (context window is details under model variant section with + signs)

They were meant to crank 2.5 to 2 million at some point though, maybe waiting now till 3?

Not a fork opencode is a from scratch project

Claude code is completely closed source and even DMCA’d people reverse engineering it.

https://techcrunch.com/2025/04/25/anthropic-sent-a-takedown-...

I use opencode and never experienced it going through tokens at the speed I experienced Gemini CLI go through.

Cant speak to qwen or crush as I have not used them

The image model is literally the same model

I’d just use sst/opencode if using other models (I use it for Claude through Claude pro subscription too)

There’s something beautiful about creating a drop in replacement for something that improves performance substantially.

ScyllaDB comes to mind

When I ask it do to do something in cursor it goes full sherlock thinking about every possible outcome.

Just claude 4 sonnet with thinking just has a bit think then does it

The playwright MCP

The orgs investing millions usually aren’t looking to invest small or even profitable most the time.

If there’s a good growth story contact people who write smaller checks who would also be interested in using the product.

Would you be okay for Google to warn anyone when entering a website on chrome that they are taking a risk because they offer payment options other than Google pay?

Its about freedom and openness of platforms.

‘It’s fine because people will ignore it’ doesn’t make something okay.