🔄 In a Training Loop

Ben Kelly PRO

YellowjacketGames

17 95 71

Shekswess's profile picture

Gipsy727's profile picture

Zenfade2's profile picture

manacasterben

AI & ML interests

None yet

Recent Activity

repliedto danielhanchen's post 2 days ago

DeepSeek-V4-Flash can now run 2× faster locally with DSpark! ⚡️ DSpark enables V4-Flash-0731 GGUFs to generate ~1.4–2× faster with no accuracy change. DeepSeek-V4-Flash-0731 can reach at 120 tokens/s. GGUFs: https://huggingface.co/unsloth/DeepSeek-V4-Flash-0731-GGUF Guide: https://unsloth.ai/docs/models/deepseek-v4

liked a model 19 days ago

BadToBest/EchoMimicV3

upvoted a paper about 1 month ago

UniMaia: Steering Chess Policies with Language for Human-like Play

View all activity

Organizations

YellowjacketGames 's collections 18

[Models] YJ Audio Stack

Curated

fishaudio/s2-pro

Text-to-Speech • 5B • Updated Mar 11 • 37.6k • 1.45k
CohereLabs/cohere-transcribe-03-2026

Automatic Speech Recognition • 2B • Updated 14 days ago • 119k • • 1.2k
Soul-AILab/SoulX-Singer

Text-to-Speech • Updated Mar 13 • 2.27k • 180
Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice

Text-to-Speech • 2B • Updated Jan 29 • 2.04M • 2.16k

[mixed] Chess x AI

Research directly related to Chess technology.

Lichess/standard-chess-games

Viewer • Updated Oct 16, 2025 • 7.14B • 804k • 116
Lichess/tournament-chess-games

Viewer • Updated Dec 9, 2025 • 931k • 731 • 9
Lichess/three-check-chess-games

Viewer • Updated Oct 16, 2025 • 9.41M • 344 • 2
Lichess/chess-puzzles

Viewer • Updated 14 days ago • 6.16M • 2.55k • 44

[models] RTX a6000 48gb

Models that run well on a *standalone* RTX a6000's 48gb of VRAM.

unsloth/Qwen-Image-Edit-2511-GGUF

Image-to-Image • 20B • Updated Jan 8 • 632k • 640
unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF

24B • Updated Dec 15, 2025 • 62.8k • 188
zai-org/GLM-Image

Text-to-Image • 7B • Updated Jan 15 • 3.44k • • 1.12k
nvidia/personaplex-7b-v1

Audio-to-Audio • 8B • Updated Mar 2 • 119k • • 2.86k

[models] CPU-Offload &/|| A6000x2

TPS can be as low as 1.0, seriously. its SLOW.

unsloth/GLM-4.7-GGUF

Text Generation • 358B • Updated Dec 27, 2025 • 6.01k • 225
unsloth/DeepSeek-R1-0528-GGUF

Text Generation • 671B • Updated Jun 15, 2025 • 15.1k • 202
unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF

Image-Text-to-Text • 401B • Updated Jun 18, 2025 • 12.5k • 51
unsloth/MiniMax-M2.1-GGUF

Text Generation • 229B • Updated Feb 14 • 3.69k • 195

[models] iGPU-Capable < 512mb

hey, you gotta try. any precision acceptable here, QA check the actual result quality. at your own risk, fool.

unsloth/LFM2.5-1.2B-Thinking-GGUF

Text Generation • 1B • Updated Aug 5 • 2.22k • 13

[models] Image Generation Stack

The stuff we actually use, pruned on an ongoing basis.

black-forest-labs/FLUX.2-dev

Image-to-Image • 32B • Updated Feb 17 • 390k • • 2.44k
unsloth/Qwen-Image-Edit-2511-GGUF

Image-to-Image • 20B • Updated Jan 8 • 632k • 640
zai-org/GLM-Image

Text-to-Image • 7B • Updated Jan 15 • 3.44k • • 1.12k
fal/Qwen-Image-Edit-2511-Multiple-Angles-LoRA

Image-to-Image • Updated Jan 7 • 67.8k • • 1.59k

[mixed] Scientific Method

Idea2Story: An Automated Pipeline for Transforming Research Concepts into Complete Scientific Narratives

Paper • 2601.20833 • Published Jan 28 • 134

[papers] RAG$ to Riche$

UltraRAG: A Modular and Automated Toolkit for Adaptive Retrieval-Augmented Generation

Paper • 2504.08761 • Published Mar 31, 2025 • 7
Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling

Paper • 2512.23959 • Published Dec 30, 2025 • 112
DRPG (Decompose, Retrieve, Plan, Generate): An Agentic Framework for Academic Rebuttal

Paper • 2601.18081 • Published Jan 26 • 8
nvidia/nemotron-colembed-vl-8b-v2

Visual Document Retrieval • 9B • Updated Feb 21 • 12.4k • 53

[papers] Image & Video

The Script is All You Need: An Agentic Framework for Long-Horizon Dialogue-to-Cinematic Video Generation

Paper • 2601.17737 • Published Jan 25 • 56
Advancing Open-source World Models

Paper • 2601.20540 • Published Jan 28 • 135
OpenVision 3: A Family of Unified Visual Encoder for Both Understanding and Generation

Paper • 2601.15369 • Published Jan 21 • 23
Video-As-Prompt: Unified Semantic Control for Video Generation

Paper • 2510.20888 • Published Oct 23, 2025 • 51

OpenClaw Fun!

A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)

Paper • 2602.14364 • Published Feb 16 • 23
Does Socialization Emerge in AI Agent Society? A Case Study of Moltbook

Paper • 2602.14299 • Published Feb 15 • 27
AIDev: Studying AI Coding Agents on GitHub

Paper • 2602.09185 • Published Feb 9 • 3
bartowski/nvidia_Nemotron-Cascade-2-30B-A3B-GGUF

Text Generation • 32B • Updated Mar 22 • 7.34k • 47

[papers] Gameplay Optimization

Research papers that may contribute to a broader approach to teaching machines how to play complex strategy games beyond just Chess.

OptiMind: Teaching LLMs to Think Like Optimization Experts

Paper • 2509.22979 • Published Sep 26, 2025 • 4
LFM2 Technical Report

Paper • 2511.23404 • Published Nov 28, 2025 • 79
Zero-Overhead Introspection for Adaptive Test-Time Compute

Paper • 2512.01457 • Published Dec 1, 2025 • 3
Confidence Estimation for LLMs in Multi-turn Interactions

Paper • 2601.02179 • Published Jan 5 • 17

[models] GTX 1660 Super 6gb

The best little card under 100 euros. Full Precision vs Quants not benchmarked. This card is so much better at running inference than you realize.

MaziyarPanahi/Nemotron-Orchestrator-8B-GGUF

Text Generation • 8B • Updated Dec 6, 2025 • 36.5k • 6
unsloth/Qwen3-4B-Instruct-2507-GGUF

4B • Updated Aug 20, 2025 • 357k • 220
unsloth/SmolLM3-3B-128K-GGUF

3B • Updated Jul 8, 2025 • 4.76k • 48
unsloth/Ministral-3-3B-Reasoning-2512-GGUF

3B • Updated Dec 4, 2025 • 10.7k • 25

[models] Sub-1gb for Edge

Toaster Tier but not iGPU

lmstudio-community/LFM2.5-1.2B-Thinking-GGUF

1B • Updated Jan 20 • 445 • 2
unsloth/SmolLM3-3B-128K-GGUF

3B • Updated Jul 8, 2025 • 4.76k • 48

[models] non-EN specialists

Emphasis on Portuguese, Spanish, Vietnamese, Hebrew

lucifrrrrrrrrrr/vn-geography-8b

8B • Updated Jan 25 • 64 • 3

[mixed] ORCAssist "Work's Done!"

YellowjacketGames/orc-assist-icons

Viewer • Updated May 21 • 48 • 414 • 1
nvidia/Nemotron-Orchestrator-8B

Text Generation • 8B • Updated Dec 2, 2025 • 2.42k • • 601
allenai/SERA-8B-GA

8B • Updated Feb 3 • 174 • 15
GAIR/daVinci-Dev-72B

Text Generation • 73B • Updated Jun 7 • 53 • 8

[papers] Distillation

Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment

Paper • 2601.14249 • Published Jan 20 • 14
Fiddler: CPU-GPU Orchestration for Fast Inference of Mixture-of-Experts Models

Paper • 2402.07033 • Published Feb 10, 2024 • 20
MeepleLM: A Virtual Playtester Simulating Diverse Subjective Experiences

Paper • 2601.07251 • Published Jan 12 • 11
GameTalk: Training LLMs for Strategic Conversation

Paper • 2601.16276 • Published Jan 22 • 14

[papers] Sports Tech

VideoMaMa: Mask-Guided Video Matting via Generative Prior

Paper • 2601.14255 • Published Jan 20 • 16
Motion 3-to-4: 3D Motion Reconstruction for 4D Synthesis

Paper • 2601.14253 • Published Jan 20 • 10
Think3D: Thinking with Space for Spatial Reasoning

Paper • 2601.13029 • Published Jan 19 • 48
netflix/void-model

Video-to-Video • Updated Apr 6 • 976

[data] What a Dump!

Seriously, look at the size of these dumps! They're a huge pile of data, dumped on your doorstep!

The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text

Paper • 2506.05209 • Published Jun 5, 2025 • 66
Lichess/standard-chess-games

Viewer • Updated Oct 16, 2025 • 7.14B • 804k • 116
Lichess/chess-puzzles

Viewer • Updated 14 days ago • 6.16M • 2.55k • 44
nvidia/SAGE-10k

Updated Feb 11 • 141k • 87

[Models] YJ Audio Stack

Curated

fishaudio/s2-pro

Text-to-Speech • 5B • Updated Mar 11 • 37.6k • 1.45k
CohereLabs/cohere-transcribe-03-2026

Automatic Speech Recognition • 2B • Updated 14 days ago • 119k • • 1.2k
Soul-AILab/SoulX-Singer

Text-to-Speech • Updated Mar 13 • 2.27k • 180
Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice

Text-to-Speech • 2B • Updated Jan 29 • 2.04M • 2.16k

OpenClaw Fun!

A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)

Paper • 2602.14364 • Published Feb 16 • 23
Does Socialization Emerge in AI Agent Society? A Case Study of Moltbook

Paper • 2602.14299 • Published Feb 15 • 27
AIDev: Studying AI Coding Agents on GitHub

Paper • 2602.09185 • Published Feb 9 • 3
bartowski/nvidia_Nemotron-Cascade-2-30B-A3B-GGUF

Text Generation • 32B • Updated Mar 22 • 7.34k • 47

[mixed] Chess x AI

Research directly related to Chess technology.

Lichess/standard-chess-games

Viewer • Updated Oct 16, 2025 • 7.14B • 804k • 116
Lichess/tournament-chess-games

Viewer • Updated Dec 9, 2025 • 931k • 731 • 9
Lichess/three-check-chess-games

Viewer • Updated Oct 16, 2025 • 9.41M • 344 • 2
Lichess/chess-puzzles

Viewer • Updated 14 days ago • 6.16M • 2.55k • 44

[papers] Gameplay Optimization

Research papers that may contribute to a broader approach to teaching machines how to play complex strategy games beyond just Chess.

OptiMind: Teaching LLMs to Think Like Optimization Experts

Paper • 2509.22979 • Published Sep 26, 2025 • 4
LFM2 Technical Report

Paper • 2511.23404 • Published Nov 28, 2025 • 79
Zero-Overhead Introspection for Adaptive Test-Time Compute

Paper • 2512.01457 • Published Dec 1, 2025 • 3
Confidence Estimation for LLMs in Multi-turn Interactions

Paper • 2601.02179 • Published Jan 5 • 17

[models] RTX a6000 48gb

Models that run well on a *standalone* RTX a6000's 48gb of VRAM.

unsloth/Qwen-Image-Edit-2511-GGUF

Image-to-Image • 20B • Updated Jan 8 • 632k • 640
unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF

24B • Updated Dec 15, 2025 • 62.8k • 188
zai-org/GLM-Image

Text-to-Image • 7B • Updated Jan 15 • 3.44k • • 1.12k
nvidia/personaplex-7b-v1

Audio-to-Audio • 8B • Updated Mar 2 • 119k • • 2.86k

[models] GTX 1660 Super 6gb

The best little card under 100 euros. Full Precision vs Quants not benchmarked. This card is so much better at running inference than you realize.

MaziyarPanahi/Nemotron-Orchestrator-8B-GGUF

Text Generation • 8B • Updated Dec 6, 2025 • 36.5k • 6
unsloth/Qwen3-4B-Instruct-2507-GGUF

4B • Updated Aug 20, 2025 • 357k • 220
unsloth/SmolLM3-3B-128K-GGUF

3B • Updated Jul 8, 2025 • 4.76k • 48
unsloth/Ministral-3-3B-Reasoning-2512-GGUF

3B • Updated Dec 4, 2025 • 10.7k • 25

[models] CPU-Offload &/|| A6000x2

TPS can be as low as 1.0, seriously. its SLOW.

unsloth/GLM-4.7-GGUF

Text Generation • 358B • Updated Dec 27, 2025 • 6.01k • 225
unsloth/DeepSeek-R1-0528-GGUF

Text Generation • 671B • Updated Jun 15, 2025 • 15.1k • 202
unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF

Image-Text-to-Text • 401B • Updated Jun 18, 2025 • 12.5k • 51
unsloth/MiniMax-M2.1-GGUF

Text Generation • 229B • Updated Feb 14 • 3.69k • 195

[models] Sub-1gb for Edge

Toaster Tier but not iGPU

lmstudio-community/LFM2.5-1.2B-Thinking-GGUF

1B • Updated Jan 20 • 445 • 2
unsloth/SmolLM3-3B-128K-GGUF

3B • Updated Jul 8, 2025 • 4.76k • 48

[models] iGPU-Capable < 512mb

hey, you gotta try. any precision acceptable here, QA check the actual result quality. at your own risk, fool.

unsloth/LFM2.5-1.2B-Thinking-GGUF

Text Generation • 1B • Updated Aug 5 • 2.22k • 13

[models] non-EN specialists

Emphasis on Portuguese, Spanish, Vietnamese, Hebrew

lucifrrrrrrrrrr/vn-geography-8b

8B • Updated Jan 25 • 64 • 3

[models] Image Generation Stack

The stuff we actually use, pruned on an ongoing basis.

black-forest-labs/FLUX.2-dev

Image-to-Image • 32B • Updated Feb 17 • 390k • • 2.44k
unsloth/Qwen-Image-Edit-2511-GGUF

Image-to-Image • 20B • Updated Jan 8 • 632k • 640
zai-org/GLM-Image

Text-to-Image • 7B • Updated Jan 15 • 3.44k • • 1.12k
fal/Qwen-Image-Edit-2511-Multiple-Angles-LoRA

Image-to-Image • Updated Jan 7 • 67.8k • • 1.59k

[mixed] ORCAssist "Work's Done!"

YellowjacketGames/orc-assist-icons

Viewer • Updated May 21 • 48 • 414 • 1
nvidia/Nemotron-Orchestrator-8B

Text Generation • 8B • Updated Dec 2, 2025 • 2.42k • • 601
allenai/SERA-8B-GA

8B • Updated Feb 3 • 174 • 15
GAIR/daVinci-Dev-72B

Text Generation • 73B • Updated Jun 7 • 53 • 8

[mixed] Scientific Method

Idea2Story: An Automated Pipeline for Transforming Research Concepts into Complete Scientific Narratives

Paper • 2601.20833 • Published Jan 28 • 134

[papers] Distillation

Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment

Paper • 2601.14249 • Published Jan 20 • 14
Fiddler: CPU-GPU Orchestration for Fast Inference of Mixture-of-Experts Models

Paper • 2402.07033 • Published Feb 10, 2024 • 20
MeepleLM: A Virtual Playtester Simulating Diverse Subjective Experiences

Paper • 2601.07251 • Published Jan 12 • 11
GameTalk: Training LLMs for Strategic Conversation

Paper • 2601.16276 • Published Jan 22 • 14

[papers] RAG$ to Riche$

UltraRAG: A Modular and Automated Toolkit for Adaptive Retrieval-Augmented Generation

Paper • 2504.08761 • Published Mar 31, 2025 • 7
Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling

Paper • 2512.23959 • Published Dec 30, 2025 • 112
DRPG (Decompose, Retrieve, Plan, Generate): An Agentic Framework for Academic Rebuttal

Paper • 2601.18081 • Published Jan 26 • 8
nvidia/nemotron-colembed-vl-8b-v2

Visual Document Retrieval • 9B • Updated Feb 21 • 12.4k • 53

[papers] Sports Tech

VideoMaMa: Mask-Guided Video Matting via Generative Prior

Paper • 2601.14255 • Published Jan 20 • 16
Motion 3-to-4: 3D Motion Reconstruction for 4D Synthesis

Paper • 2601.14253 • Published Jan 20 • 10
Think3D: Thinking with Space for Spatial Reasoning

Paper • 2601.13029 • Published Jan 19 • 48
netflix/void-model

Video-to-Video • Updated Apr 6 • 976

[papers] Image & Video

The Script is All You Need: An Agentic Framework for Long-Horizon Dialogue-to-Cinematic Video Generation

Paper • 2601.17737 • Published Jan 25 • 56
Advancing Open-source World Models

Paper • 2601.20540 • Published Jan 28 • 135
OpenVision 3: A Family of Unified Visual Encoder for Both Understanding and Generation

Paper • 2601.15369 • Published Jan 21 • 23
Video-As-Prompt: Unified Semantic Control for Video Generation

Paper • 2510.20888 • Published Oct 23, 2025 • 51

[data] What a Dump!

Seriously, look at the size of these dumps! They're a huge pile of data, dumped on your doorstep!

The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text

Paper • 2506.05209 • Published Jun 5, 2025 • 66
Lichess/standard-chess-games

Viewer • Updated Oct 16, 2025 • 7.14B • 804k • 116
Lichess/chess-puzzles

Viewer • Updated 14 days ago • 6.16M • 2.55k • 44
nvidia/SAGE-10k

Updated Feb 11 • 141k • 87