i come in peace

akirathegreatest

2 7

AI & ML interests

None yet

Recent Activity

reacted to SeaWolf-AI's post with 👍 3 days ago

💻 Data-center AI, now on a laptop: POCKET-Darwin-180B We're releasing a 4-bit GGUF build of Darwin-180B-RSI, #1 on seven official Hugging Face leaderboards (self-reported), that runs without a GPU. 📦 360 GB → 111 GB (4-bit GGUF, 4 files) 🖥️ No GPU: one server CPU (16 threads) at 18.4–21.0 tokens/s 💻 RTX 5060 laptop (8 GB VRAM) + 32 GB RAM: 4.17 tokens/s 🧊 128 GB mini PC: whole model in memory, no GPU needed 🎯 MMLU-Pro, 2,000 questions, paired: original 87.65% = 4-bit 87.65% How? · Only ~3B of 180B parameters are active per token (10 of 512 experts) · llama.cpp streams just the needed experts from SSD, so 32 GB RAM is enough · Graft quantization: we took the proven Unsloth UD-Q4_K_XL base build and swapped in only the 300 tensors our RSI training changed (300/300 verified) Under the hood is Model-level Recursive Self-Improvement. The model solves verifiable problems, keeps only its own solutions that check out as correct, and trains on them. No human-written solutions or reasoning traces. Built for teams that can't send data to an external cloud (defense, finance, public sector) to run a top-tier model fully offline. 📝 Article: https://huggingface.co/blog/FINAL-Bench/data-center-ai-now-on-a-laptop-pocket-darwin-180b 🤗 Model: https://huggingface.co/FINAL-Bench/POCKET-Darwin-180B-GGUF 🧬 Original: https://huggingface.co/FINAL-Bench/Darwin-180B-RSI #Darwin #RSI #GGUF #llamacpp #OnDevice #MoE

upvoted a paper 2 months ago

A Frozen 12B Beats Frontier Models on Verified Work: 100% Accuracy, 0 Tokens, Bit-Exact, Forever

reacted to kasbsquall's post with 🔥 4 months ago

🔎 UX Crime Scene — every interface hides a crime. Drop a screenshot of ANY website or app, and THE INSPECTOR — a film-noir detective — works it as a crime scene: he circles each UX flaw on the real pixels, names the charge, and files a verdict with a letter grade. A UX audit that plays like a detective thriller. But the verdict is just the opening statement. Now it goes further: ⚖️ THE TRIAL — put the interface on trial. The guilty UI elements take the stand and defend themselves while the Inspector rules from the evidence. 🖼️ THE RECONSTRUCTION — one click and FLUX.2 Klein rebuilds the worst element FIXED, live. Before/after, on the real pixels. 🔊 THE VOICE — hear the verdict read aloud (Kokoro, local, no keys). 🚨 MOST WANTED — a public rogues' gallery. Book your case onto a shared board where the city's worst interfaces are ranked by their crimes. Booked by the public. Three small models, all on Modal (scale-to-zero), none over 32B: 👁️ Qwen2.5-VL-7B (vision agent) · 🖼️ FLUX.2 Klein (reconstruction) · 🔊 Kokoro-82M (voice) 📊 Human-graded: 84% grounding / 92% valid charges. ▶️ Trailer: https://youtu.be/6u58YIEPrkA 📹 Full walkthrough: https://youtu.be/WyQbY0XJ_9E 🕵️ Try it: https://huggingface.co/spaces/build-small-hackathon/ux-crime-scene Built solo for #BuildSmallHackathon (Gradio × Hugging Face). Open the case — the Inspector is waiting.

View all activity

Organizations

None yet

reacted to SeaWolf-AI's post with 👍 3 days ago

Post

6160

💻 Data-center AI, now on a laptop: POCKET-Darwin-180B

We're releasing a 4-bit GGUF build of Darwin-180B-RSI, #1 on seven official Hugging Face leaderboards (self-reported), that runs without a GPU.

📦 360 GB → 111 GB (4-bit GGUF, 4 files)
🖥️ No GPU: one server CPU (16 threads) at 18.4–21.0 tokens/s
💻 RTX 5060 laptop (8 GB VRAM) + 32 GB RAM: 4.17 tokens/s
🧊 128 GB mini PC: whole model in memory, no GPU needed
🎯 MMLU-Pro, 2,000 questions, paired: original 87.65% = 4-bit 87.65%

How?
· Only ~3B of 180B parameters are active per token (10 of 512 experts)
· llama.cpp streams just the needed experts from SSD, so 32 GB RAM is enough
· Graft quantization: we took the proven Unsloth UD-Q4_K_XL base build and swapped in only the 300 tensors our RSI training changed (300/300 verified)

Under the hood is Model-level Recursive Self-Improvement. The model solves verifiable problems, keeps only its own solutions that check out as correct, and trains on them. No human-written solutions or reasoning traces.

Built for teams that can't send data to an external cloud (defense, finance, public sector) to run a top-tier model fully offline.

📝 Article: https://huggingface.co/blog/FINAL-Bench/data-center-ai-now-on-a-laptop-pocket-darwin-180b
🤗 Model: FINAL-Bench/POCKET-Darwin-180B-GGUF
🧬 Original: FINAL-Bench/Darwin-180B-RSI

#Darwin #RSI #GGUF #llamacpp #OnDevice #MoE

30 replies

upvoted a paper 2 months ago

A Frozen 12B Beats Frontier Models on Verified Work: 100% Accuracy, 0 Tokens, Bit-Exact, Forever

Paper • 2607.23806 • Published Jul 26 • 8

reacted to kasbsquall's post with 🔥 4 months ago

Post

4193

🔎 UX Crime Scene — every interface hides a crime.

Drop a screenshot of ANY website or app, and THE INSPECTOR — a film-noir detective — works it as a crime scene: he circles each UX flaw on the real pixels, names the charge, and files a verdict with a letter grade. A UX audit that plays like a detective thriller.

But the verdict is just the opening statement. Now it goes further:

⚖️ THE TRIAL — put the interface on trial. The guilty UI elements take the stand and defend themselves while the Inspector rules from the evidence.
🖼️ THE RECONSTRUCTION — one click and FLUX.2 Klein rebuilds the worst element FIXED, live. Before/after, on the real pixels.
🔊 THE VOICE — hear the verdict read aloud (Kokoro, local, no keys).
🚨 MOST WANTED — a public rogues' gallery. Book your case onto a shared board where the city's worst interfaces are ranked by their crimes. Booked by the public.

Three small models, all on Modal (scale-to-zero), none over 32B:
👁️ Qwen2.5-VL-7B (vision agent) · 🖼️ FLUX.2 Klein (reconstruction) · 🔊 Kokoro-82M (voice)

📊 Human-graded: 84% grounding / 92% valid charges.

▶️ Trailer: https://youtu.be/6u58YIEPrkA
📹 Full walkthrough: https://youtu.be/WyQbY0XJ_9E
🕵️ Try it: build-small-hackathon/ux-crime-scene

Built solo for #BuildSmallHackathon (Gradio × Hugging Face). Open the case — the Inspector is waiting.

1 reply

reacted to sergiopaniego's post with 🔥 10 months ago

Post

2904

ICYMI, you can fine-tune open LLMs using Claude Code

just tell it:
“Fine-tune Qwen3-0.6B on open-r1/codeforces-cots”

and Claude submits a real training job on HF GPUs using TRL.

it handles everything:
> dataset validation
> GPU selection
> training + Trackio monitoring
> job submission + cost estimation
when it’s done, your model is on the Hub, ready to use

read more about the process: https://huggingface.co/blog/hf-skills-training