Senior Producer Virtual production, Real time graphics. data visualization. Crypto. VFX
somewhere hot
Joined September 2008
- Tweets10K
- Following2.5K
- Followers765
- Likes52.8K
AussieAck ττ retweeted
around the month of March 2026, one subnet training a small model in a decentralized setup sent $TAO to 350$+ until the subnet owner rugged and fleed.
this subnet was trash, barely scratching the surface of what we currently have cooking
the difference now is that it has not yet been shilled on a stage with Jensen Huang noding in passive approval to it
position wisely
parallax.chutes.ai/
AussieAck ττ retweeted
The push for regulatory capture and AI cabalism is accelerating. This, in large part, is why I'm building parallax. True P2P unstoppable, uncensorable AI systems aren't just neat, they're mandatory.
This little single DGX spark "datacenter" in my forge with parallax can train (with other nodes) an 80b parameter model.
- two solar blankets
- ecoflow battery
- lapdock for kvm
- starlink internet
- dgx spark
Total input from solar ~600w (800w max)
Total power draw from spark/starlink/etc. < 200w
No datacenters, no water supplies, no external power grid, no multi gigabit link.
We must design AI systems that are as resilient as the humble cockroach.
Why do we all appreciate BTC? Because it's a mathematically certain sovereign ledger that can't be inflated away or centrally controlled.
Current AI represent the greatest concentration of wealth, power, and information (IP theft? blackmail? or simply ads and behavior control?) the world has ever seen... As BTC did to finances, we must do to AI.
Chutes + parallax is a step forward towards tearing down this moat.
AussieAck ττ retweeted
Minimax H3 😀
-Advanced Prompting style
with micro-expressions + emotions
using tags and context for natural detailed output
full details in comment
AussieAck ττ retweeted
Dashboard is a quick work in progress, but for visibility into a run here ya go!
parallax.chutes.ai/
This is around $11/b tokens, insane actually. MFU also insane. The whole thing, pretty legendary, and inference...
Using a few nodes from @lium_io also!
AussieAck ττ retweeted
Hussam Ebu Safieh, "İsrail'in rehineler için ölüm cezası" ile öldürülecek olan Filistinli doktorlardan biridir (diğer 95 doktor arasında).
Onu öldürmelerine izin verme.
Bunu yeniden yayınlayın.
AussieAck ττ retweeted
"Israel" ha convertido la Franja de Gaza en el lugar con más niños amputados del mundo, más de 40.000 niños palestinos han sido heridos, perdiendo extremidades, en 3 años, en una media de 36 niños amputados al día.
Elegir el silencio ante la barbarie, es elegir la barbarie.
AussieAck ττ retweeted
FFmpeg Skill Update 🎬
Ver.0.9.2→Ver.0.12.0
「この動画を9:16にして」「無音カットして」「字幕入れて」
に加えて、今はこんな指示もできます。
「この部分だけピクセル単位でクロップして」
「この写真、ズームしながらクリップにして」(Ken Burns風)
「動画を回転/反転して」
「この動画にロゴ動画を合成して」「グリーンバックを抜いて」
「この動画、逆再生にして」
「手ブレを直して」
「連番の画像から動画作って」
「単色/グラデーションの背景クリップ作って」
「劣化なしでそのまま書き出して」
「字幕は画質そのままで、後から切り替えられる形で入れて」
「ハイライトは音量じゃなくて長さで選んで」
「AI解析用に軽い解像度でプレビュー書き出して」
できることが21→28個のツールに増えました。
🔍 AIが「わかった上で」処理するように
・「今この環境でこのツールが実際に使えるか」を実行前に事前チェック
・GPUエンコーダー/フォント代替の有無も把握
・複数の音声トラックを持つファイル(吹き替え・別言語版など)で、使うトラックをほぼ全ツールで指定可能に
・字幕トラックの言語などの詳細も読み取れるように
・字幕タイミングをタイムコード(hh:mm:ss:ff)形式でも指定可能に
・9:16化などのクロップで、切り取る位置(左寄せ/右寄せなど)も指定可能に
・音声を再エンコードするかしないか、各ツールが正直に申告するように
✅ 「壊れない・嘘をつかない」を徹底
・書き出しが本当に成功したか毎回確認してから「完了」と報告、失敗時のゴミファイルも残さない
・元の動画ファイルを誤って上書き・破損させない安全策
・失敗した時、原因とやり直せるかをAIが判別
・色変換で再エンコードが必要になった場合も、字幕や追加の音声トラックをできる限り残すように
・同期のズレを検出した時、信頼度が低い場合はAIに警告
・最新FFmpeg / Windows環境でも安定動作(Windowsのパス周りのバグ修正含む)
🛠️ 開発・保守の仕組みも強化
・「AIがして良いこと/判断すべきでないこと」を明文化(勝手にAIが「見た目の良さ」を判断しない設計)
・生FFmpegコマンドへの逃げ道を塞ぐテストを追加
・ドキュメントと実際のコードの挙動がズレたら自動テストで検知
・OSSとしてのContributing手順を整備、バージョンタグを打つとGitHub Releaseが自動生成される仕組みを追加
・GPU対応について「対応してるとは言うが動作保証はしない」ことを正直に明記
Claude Code / Cursor / Codex対応、MCPあり、ローカル処理でクラウド不要は変わらず。
Current: v0.12.0
github.com/kajisho5/ffmpeg-s…
#FFmpeg #AIVideo #AI #MCP #ClaudeCode #OpenSource
AussieAck ττ retweeted
We're making the first movie entirely with GPT-6 Astra computer-use.
We'll show the process below.
AussieAck ττ retweeted
Minimax H3 running locally.
~65GB of models.
12GB VRAM. RTX 4070 Ti.
100% Text-to-Video.
100% ComfyUI.
Pass 1 → 960×544
Pass 2 → 1920×1088
One continuous loop built from multiple context-connected generations.
No cloud. No API.
CONTROL IS THE KEY.
LOCAL IS THE WAY.
AussieAck ττ retweeted
NVIDIA B300 servers retail around $350k min, crazy power/cooling requirements/etc. (bad maths) per GPU (8) is $43,750, if you can even find them (narrator: you can't) then pay for collocation etc... 9 PFLOPS dense FP4.
DGX Spark: $4700, plug in anywhere, 1 PFLOP (fp4 sparse).
B300 9ish PFLOP dense FP4 super expensive impossible to collocate anywhere etc. etc., $4861.11 per PFLOP.
DGX Spark $4700 plug in any 15 amp outline in any house, $4700/PFLOP (but sparse).
Memory is slower, so I guess use less of it/less manipulation of it...
Yep, math checks out - pretraining on decentralized DGX Spark with native sparse FP4 (checks notes: yes we're using ternary weights which slot perfectly into sparse FP4 compute) should be a thing.
Time to test it out...
AussieAck ττ retweeted
Reminder:
BOMBSHELL: Leaked footage from Israeli Airforce proves it was the IDF that took out its own citizens with Apache Helicopters on October 7 at the Nova music festival….
Israel has now admitted 28 Apache helicopters discharged their ammunition and had to be reloaded…
AussieAck ττ retweeted
A 0.6B speech model should not sound this capable.
Today, we’re open-sourcing Audio8-TTS Preview:
11 languages.
Zero-shot voice cloning.
A bundled 44.1 kHz codec.
Apache 2.0.
Small model. Serious speech.
Thread ↓
AussieAck ττ retweeted
THIS COULD BE A $50,000 BLENDER COMMERCIAL.
The scene looks like it came from a full production studio: detailed environments, cinematic camera work, realistic materials, complex lighting, simulations, and dozens of small elements all moving together.
A high-end brand campaign with this level of custom 3D work could easily reach $30,000–$70,000.
The bigger opportunity is what happens when Blender workflows start becoming easier to control with MCP.
You no longer need to spend hours clicking through every menu just to test an idea. You can move faster, build more variations, and turn concepts into finished visuals without needing a large 3D team.
That creates a real business opportunity.
Brands need product films. Startups need launch visuals. Agencies need cinematic content that doesn't look like another AI-generated template.
The people learning Blender + MCP now aren't just making cool renders.
They're building a new kind of creative service.
And there is serious money in it.
AussieAck ττ retweeted
ANTHROPIC ENGINEER: MOST PEOPLE ARE STILL PROMPTING ONE STEP AT A TIME.
An Anthropic engineer says the gap is not the model. "Fable 5 is already smarter than we know how to use. The bottleneck was never the AI. It's you."
20 minutes, no extra spend, showing how to build something that improves itself instead of needing you to prompt every single step forever.
Watch it, then save the guide below.
AussieAck ττ retweeted
EVERY TIME WE UPLOAD THE VIDEO AND IT GETS DELETED, WE'LL JUST KEEP UPLOADING IT. 🤣👌
