The Studio: Taming the Chaos. Crafting the Simple.
Running eGPU spaghetti monsters to build ecosystem models, code, and craft media. Current workflow status: localmaxxing, 500M+ tokens and climbing!
#GameDev #OpenSource #CreativeCoding #eGPU #Localmaxxing
Running eGPU spaghetti monsters to build ecosystem models, code, and craft media. Current workflow status: localmaxxing, 500M+ tokens and climbing!
#GameDev #OpenSource #CreativeCoding #eGPU #Localmaxxing
June 8, 2026 at 1:43 PM
The Studio: Taming the Chaos. Crafting the Simple.
Running eGPU spaghetti monsters to build ecosystem models, code, and craft media. Current workflow status: localmaxxing, 500M+ tokens and climbing!
#GameDev #OpenSource #CreativeCoding #eGPU #Localmaxxing
Running eGPU spaghetti monsters to build ecosystem models, code, and craft media. Current workflow status: localmaxxing, 500M+ tokens and climbing!
#GameDev #OpenSource #CreativeCoding #eGPU #Localmaxxing
‘Localmaxxing is an inevitable response to tokenmaxxing. As local models improve & close the gap with frontier, more users will shift workloads to their own hardware. If half the work runs 2x faster on my laptop, I’ll take that trade every time.‘ tomtunguz.com/localmaxxing/
Localmaxxing
About half of agent tasks can run on a local 35B model. The real advantage isn't cost or privacy — it's latency. 2.1x faster means more iteration cycles per session.
tomtunguz.com
May 11, 2026 at 7:52 PM
‘Localmaxxing is an inevitable response to tokenmaxxing. As local models improve & close the gap with frontier, more users will shift workloads to their own hardware. If half the work runs 2x faster on my laptop, I’ll take that trade every time.‘ tomtunguz.com/localmaxxing/
Localmaxxing #devopsish tomtunguz.com/localm...
May 14, 2026 at 11:51 PM
Localmaxxing #devopsish tomtunguz.com/localm...
RT @LottoLabs: All you have to do is go to http://localmaxxing.com and just see what models people are actually running and what hardware you actually need, no more guessing no more speculation
Localmaxxing - Local LLM Inference Benchmarks
Community-sourced benchmarks for local LLM inference. Compare tokens/sec across GPUs, Apple Silicon, CPUs, and engines like llama.cpp, Ollama, and LM Studio.
localmaxxing.com
July 8, 2026 at 1:31 AM
RT @LottoLabs: All you have to do is go to http://localmaxxing.com and just see what models people are actually running and what hardware you actually need, no more guessing no more speculation
localmaxxing | tomasz tunguz https://tomtunguz.com/localmaxxing/
September 4, 2026 at 10:29 AM
localmaxxing | tomasz tunguz https://tomtunguz.com/localmaxxing/
guessing is over thanks to localmaxxing
July 8, 2026 at 6:52 AM
guessing is over thanks to localmaxxing
OpenClaw landed on @huggingface local apps 🦞🤝🤗
1. Pick any GGUF/MLX model on hf
2. Copy the openclaw onboard setup
3. Volla you've got a tool-calling agent running fully local. no cloud, no keys, no one watching.
Get your claw localmaxxing. resistance is futile 🦞
1. Pick any GGUF/MLX model on hf
2. Copy the openclaw onboard setup
3. Volla you've got a tool-calling agent running fully local. no cloud, no keys, no one watching.
Get your claw localmaxxing. resistance is futile 🦞
August 5, 2026 at 7:54 AM
OpenClaw landed on @huggingface local apps 🦞🤝🤗
1. Pick any GGUF/MLX model on hf
2. Copy the openclaw onboard setup
3. Volla you've got a tool-calling agent running fully local. no cloud, no keys, no one watching.
Get your claw localmaxxing. resistance is futile 🦞
1. Pick any GGUF/MLX model on hf
2. Copy the openclaw onboard setup
3. Volla you've got a tool-calling agent running fully local. no cloud, no keys, no one watching.
Get your claw localmaxxing. resistance is futile 🦞
Mike Key氏がClaude Code向けスキルを近日公開予定、マシン上の全LLMをエージェント型コーディング向けか速度向けに体系的に調整する
既存のclaude-toolkitとLocalMaxxingで計測基盤を持つ作者が、設定の属人化を手順化する動き
#upppp
https://upppp.jp/ai-usecase/20260628-192619/
既存のclaude-toolkitとLocalMaxxingで計測基盤を持つ作者が、設定の属人化を手順化する動き
#upppp
https://upppp.jp/ai-usecase/20260628-192619/
upppp.jp
June 28, 2026 at 10:26 AM
Mike Key氏がClaude Code向けスキルを近日公開予定、マシン上の全LLMをエージェント型コーディング向けか速度向けに体系的に調整する
既存のclaude-toolkitとLocalMaxxingで計測基盤を持つ作者が、設定の属人化を手順化する動き
#upppp
https://upppp.jp/ai-usecase/20260628-192619/
既存のclaude-toolkitとLocalMaxxingで計測基盤を持つ作者が、設定の属人化を手順化する動き
#upppp
https://upppp.jp/ai-usecase/20260628-192619/
