@opencode Remote v1.3.6 is out.
This release improves global session discovery, activity sorting, directory-scoped sessions, new-session validation, and model picker spacing.
Signed Android APK available here:
https://t.co/QdRJhNa5gg
@LucianoLicelli Non è un server, ma un ottimo PC sì: compatto, silenziosissimo, dai consumi ridottissimi e praticamente unico nella sua fascia.
Nel mondo NVIDIA, il solo costo energetico delle inferenze locali può superare quello di un uso regolare dei principali LLM più potenti via API.
@LucianoLicelli Per produzione servono crosti di un altro ordine di grandezza. Sia come costi per inferenze sia come costi hardware. AMD in questo momento ha un target consumer e non server o produzione. E come consumer meglio un Halo Strix di una RTX che tra rumore e consumo non ha senso.
@LucianoLicelli Esattamente. Peraltro, i nuovi sistemi basati sul Ryzen AI MAX+ 495 di AMD non aumentano la bandwidth: offrono principalmente più RAM e una CPU leggermente più veloce. Va però detto che i consumi restano estremamente contenuti se confrontati con sistemi NVIDIA RTX o Blackwell.
Coming in OpenCode Desktop v2
10x faster performance (measured)
I’m not kidding!
I used a 30x cpu slowdown for my baseline at 4fps on an already weak machine, getting to approximately 45fps.
- Migrated to tanstack virtual
- Removed flicker, scroll (unintentional) jumping and lag while scrolling
- Updated to the latest Pierre diffs version and shiki versions for free gains
- Offloaded Shiki to a worker thread (off UI thread!!)
- A number of performance tricks based on the hot path append only assumptions
- Generic performance trace cpu time optimisations
🚀 OpenCode Remote Android v1.3.0 is now available.
Monitor your OpenCode sessions, send prompts, and stay connected to your AI workflows directly from your phone.
https://t.co/o9MZRgESqO
#OpenCode#Android#OpenSource#AI
@opencode I tried to let it code and fix bugs of my OpenCode Android app but it produced other bugs and mess to the codebase. Deepseek v4 Flash is a lot better.
https://t.co/RvNQm166SK
3 prompts.
40 minutes.
Fully local.
Vibe codee a web-based Pacman game on a 96GB Halo Strix mini PC using:
- tweaked llama.cpp
- @AMD Halo Strix
- @Alibaba_Qwen Qwen 3.6 35B
- @opencode_ai for vibe coding
No cloud GPUs. No API costs.
Just local agentic coding.
#LocalAI#LLM
@Pranit I wouldn’t risk too much when Open Source models are improving this fast.
The gap between coding well and coding better will eventually shrink.
And if an Open Source model is “good enough” at 1/15 of the cost… that changes everything.
@tonysimons_@JoelDeTeves Hermes OpenViking plugin makes OpenViking not completely reliable. My agent had to patch the plugin multiple times, sent a PR with the fixes to Hermes. I have a local install of OpenViking with local models for embeddings and OpenRouter for retrieval.
Anthropic analyzed 2M Claude chats. The takeaway:
AI usage today looks more like collaboration than replacement.
The real risk might be junior roles: if AI does the entry-level work, how will the next generation gain experience?
Curious about your thoughts.
#AI#FutureOfWork
@billyrestey Nice! The only glitch I noticed is that he has a left hand instead of the right one when clicking the button. Apart from this it's amazing.
Just shipped OpenCode Remote app v1.0.0 🚀
A mobile-first open source companion app to control your OpenCode server from phone/desktop.
Repo: https://t.co/RvNQm166SK
#OpenSource#Android#WebDev#AI#OpenCode