27 Delphi midtraining endpoint checkpoints (K=0.20): 9 scales x 3 mixes. Dense Qwen3, Llama-3 tok. marin#6279.
LAION eV
non-profit
AI & ML interests
open multi-modal foundation models and datasets for their creation; scaling laws, model evaluation; fully local, sovereign model deployment, personalized assistants and open local agentic systems
Recent Activity
View all activity
Organization Card
models 847
laion/delphi-1e21-p50m50-9p25b-lr0_83-f9edd2-magpie_lr1e5-sft
Text Generation • 142k • Updated
laion/delphi-2e19-837Mparams-3.6Btokens-base-wc386k_lr1e5-sft
Text Generation • 0.8B • Updated
laion/delphi-9e18-550Mparams-2.9Btokens-base-wc386k_lr1e5-sft
Text Generation • 0.6B • Updated
laion/delphi-3e18-447Mparams-1.2Btokens-base-wc386k_lr1e5-sft
Text Generation • 0.4B • Updated
laion/speech-artifact-detectors
Updated
laion/stageD-thinkbudget
8B • Updated • 1
laion/stageB-channel-80-8B
8B • Updated • 14
laion/delphi-1e21-p33m67-9p25b-lr0_67-9cf8da-magpie_lr1e5-sft
Text Generation • 142k • Updated • 4
laion/stageB-channel
8B • Updated • 1
laion/voiceclap-commercial
Feature Extraction • 0.1B • Updated • 17
datasets 445
laion/tts-samples-with-artifacts
Viewer • Updated • 6k • 18
laion/BVD-I-300M-URLs
Preview • Updated • 37 • 1
laion/GLM-4.7-swesmith-oracle_verified-complete-lt32k
Viewer • Updated • 7.32k • 28
laion/swegym-tasks-patched-validated-v5
Viewer • Updated • 2.44k • 35
laion/swegym-tasks-patched-validated-v4
Viewer • Updated • 2.44k • 25
laion/swegym-tasks-patched-validated-v3
Viewer • Updated • 2.44k • 27
laion/character-voices
Viewer • Updated • 2.18k • 1.42k
laion/soundscape-bench
Viewer • Updated • 200 • 419 • 1
laion/llama-nemotron-science-reasoning-on-le3000tok-100k-canonical-think
Viewer • Updated • 100k • 27
laion/llama-nemotron-science-reasoning-on-le3000tok-100k
Viewer • Updated • 100k • 39