Jerell

Creator of Spectra-Terminal · Inkrid Universal DevToolbox

[Jerell @cwfox67]

@_FORAB 馬斯克自己也沒好到哪去，過度的超高科技集中在一個人手上更危險。

[Jerell @cwfox67]

@liming_h 無感，我反而覺得 GPT $20 很划算，可能是因為我用 pi 的關係？ (不是透過 codex cli)，最近一直在考慮停掉 Claude Code $100訂閱換過來..

[Jerell @cwfox67]

@NVIDIAAI We need a high fecient coding model. 🤣 especially for GB10 chip.

[Jerell @cwfox67]

@jimmy_su @vapor0x 16gb ram 應該跑的動 TranslateGemma 4b/12b 版本..
另外想許願自訂provider和自訂快捷鍵. 😆

[Jerell @cwfox67]

@jimmy_su @vapor0x 之前有出個 translategemma 還不錯喔

[Jerell @cwfox67]

Qwen3.6-35B-A3B-UD-MLX-4bit was surrprised me, running on mac mini m4 pro (48gb ram), with omlx backend + openclaude cli (also tested with pi).

Tested on my real work codebase, it's fully run with a bug fix planning/feature planning. but crashed on tools_use.

and just performed around 3xx PP/ 30~30 tok/s.

[Jerell @cwfox67]

@ollama @ErRahul337 hope ollama can make owned coding cli, I am stick around claude code/pi/opencode for my real life working, but I hate always needs to patch config files for all of them, just only to make them works with ollama/custom provider/official provider.

[Jerell @cwfox67]

@jun_song @Kingotdamn @bun_byte but so far OEMs GB10 model are way cheaper than a Mac Studio M4/5 max.

[Jerell @cwfox67]

@jun_song @Kingotdamn @bun_byte how about DGX Spark or other OEMs GB10 models?

[Jerell @cwfox67]

Running Qwen3.6-35B-A3B-UD-MLX-4bit around 30~34 tok/s locally on single mac mini m4 pro (48gb ram onboard) with oMLX backend (enable TurboQuant option) and claude code.

in my opinion, it obviously works and not so bad.

[Jerell @cwfox67]

@OsaurusAI OK, got it ! Thank you. 😆

[Jerell @cwfox67]

@OsaurusAI thanks ! but I can't find it. seems only can search specific provider on huggingface.

[Jerell @cwfox67]

@ClaudeDevs here is another problem I'm encountering is that I've set it to "auto effort", but Claude Code automatically sets it to the "max effort" level.

[Jerell @cwfox67]

@sergey_moloman @ClaudeDevs same here.

[Jerell @cwfox67]

@bcherny If I set it to /effort auto, will I achieve the best balance between efficiency/intelligence/token consumption?

[Jerell @cwfox67]

@jacob_posel yap, coz Opus 4.6 just act like an idiot today

[Jerell @cwfox67]

Spectra 0.2.4 is out. It's a macOS terminal built on libghostty.

- Fixed issue of Send Copy To was silently skipping same-named folders.
- Fixed issue of copying files to SMB/NAS shares failed with "Permission denied".

https://t.co/5ghMAT25Au

[Jerell @cwfox67]

Shipped my side project: Spectra — a macOS terminal that *remembers* your setup.
Splits, tabs, directories → save once as a workspace.
Next time: one click, back to flow.
No more morning re-arranging.

https://t.co/5ghMAT25Au

[Jerell @cwfox67]

@gslin 的確。