TKAI

thaki-AI's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

updated 2 collections about 1 hour ago

EmbeddingGemma 2 Edge — Quantization

Collection

Quantized EmbeddingGemma 2 text towers that query an existing BF16 index without re-embedding the corpus. • 5 items • Updated about 1 hour ago

SkillRet — Agent Skill Retrieval

Collection

Retrieval embeddings, benchmark, and index that let an LLM agent find the right skill inside a library of 1,000+ skills. • 16 items • Updated about 1 hour ago

updated 2 models about 1 hour ago

ThakiCloud/eg2-text-q4

Feature Extraction • 0.1B • Updated about 1 hour ago

ThakiCloud/SKILLRET-Edge-22M-drift-mixed

Feature Extraction • 22.7M • Updated about 1 hour ago • 66

updated a collection about 1 hour ago

EmbeddingGemma 2 Edge — Quantization

Collection

Quantized EmbeddingGemma 2 text towers that query an existing BF16 index without re-embedding the corpus. • 5 items • Updated about 1 hour ago

updated 2 models about 1 hour ago

ThakiCloud/eg2-text-hybrid-129m

Feature Extraction • 0.1B • Updated about 1 hour ago

ThakiCloud/eg2-text-hybrid-119m

Feature Extraction • 0.1B • Updated about 1 hour ago

published a model about 1 hour ago

ThakiCloud/eg2-text-hybrid-119m

Feature Extraction • 0.1B • Updated about 1 hour ago

updated a collection about 7 hours ago

EmbeddingGemma 2 Edge — Quantization

Collection

Quantized EmbeddingGemma 2 text towers that query an existing BF16 index without re-embedding the corpus. • 5 items • Updated about 1 hour ago

published a model about 7 hours ago

ThakiCloud/eg2-text-hybrid-129m

Feature Extraction • 0.1B • Updated about 1 hour ago

updated a dataset about 10 hours ago

thaki-AI/daily-paper-2026-10-08-outcome-feedback-skill-router

Updated about 10 hours ago

published a dataset about 10 hours ago

thaki-AI/daily-paper-2026-10-08-outcome-feedback-skill-router

Updated about 10 hours ago

updated 2 collections about 21 hours ago

EmbeddingGemma 2 Edge — Quantization

Collection

Quantized EmbeddingGemma 2 text towers that query an existing BF16 index without re-embedding the corpus. • 5 items • Updated about 1 hour ago

RAG-Gate — Answer / Retrieve / Stop Gates for RAG

Collection

One-token gates after retrieval, before generation: answer, retrieve more, or stop. Released after five pre-registered gates. • 5 items • Updated about 21 hours ago

published a model about 21 hours ago

ThakiCloud/eg2-text-q4

Feature Extraction • 0.1B • Updated about 1 hour ago

published a model about 22 hours ago

ThakiCloud/RAG-Gate-27B

Text Generation • 27B • Updated about 21 hours ago

updated a model about 22 hours ago

ThakiCloud/RAG-Gate-27B

Text Generation • 27B • Updated about 21 hours ago

liked a model 1 day ago

ThakiCloud/ChainCheck-Judge-Qwen3.8-27B

Text Generation • 28B • Updated 2 days ago • 497 • 2