ρrαnαm

Grit.

[ρrαnαm @pnxshy]

@RudrakshG05 @maanas_tyagi @WisprFlow have you tried handy? its open source and you can set whatever model you like

[ρrαnαm @pnxshy]

@RobKnight__ next : photoshop in terminal

[ρrαnαm @pnxshy]

@nutlope What about the ui? is it Opus or an Anthropic model? how much did that cost?

[ρrαnαm @pnxshy]

@fforres @Cloudflare you dont like origin?

[ρrαnαm @pnxshy]

Current setup:

Main: DeepSeek V4 Flash 0731 via OpenRouter

Sub: Mimo V2.5 (needs steering) + Spark 1.2 (docs/writing + testing) via OpenCode

Luna: Only for plans involving multiple interactions/lookups across different files, mainly for non-breaking changes.

[ρrαnαm @pnxshy]

@jason_cxyz @casx718 @yacineMTB its not going to be the same from today. unless you use the api on off peak hours.

[ρrαnαm @pnxshy]

@rubenhume maximum User Retention Rate

[ρrαnαm @pnxshy]

@AvinashSingh_20 why not both?

[ρrαnαm @pnxshy]

@scottymatt seems like a minimal https://t.co/77nfjI0O0m

[ρrαnαm @pnxshy]

@maanas_tyagi host glimmer 30B on that gtx and make that as your comfort space.

[ρrαnαm @pnxshy]

@1kartikkabadi1 @badlogicgames If you are primarily using Deepseek as you LLM try Reasonix

[ρrαnαm @pnxshy]

Has anyone tested the new Meta's Glimmer yet? Is it better than Qwen-3.5-27B?

https://t.co/oiIfbmXGMZ

[ρrαnαm @pnxshy]

@CodeWithAmann you get yt music, no buffering, no extensions to hog on memory and better bitrate.

[ρrαnαm @pnxshy]

@shownotover yeah now watch them refund it 🤣

[ρrαnαm @pnxshy]

@Shra_va_ni works on mine tho.

[ρrαnαm @pnxshy]

@mil000 thinkpad missing.

[Jay @jayair]

I’m sorry but a model escaping a “sandbox” says more about the sandbox than the model

[ρrαnαm @pnxshy]

@yash_yk45 @opencode Claude still has its moments.

[ρrαnαm @pnxshy]

@theo_id3 @nalinrajput23 in what tho?

[ρrαnαm @pnxshy]

@Tirth_dev22 @maanas_tyagi https se use karle