EsKa's picture

EsKa

SerialKicked

·

https://github.com/SerialKicked/Lethe-AI-Sharp/

SerialKicked

AI & ML interests

Working on Lethe AI, a modular, object‑oriented C# library that connects local or remote LLM backends to applications. Feature long term memory, backend agnostic code, web search support, and agentic tasks, all rolled into one neat C# lib.

Recent Activity

updated a collection about 3 hours ago

Untested / Unsorted models

commented on an article 4 days ago

The Dark Underbelly of Anthropic: How the "Responsible" AI Pioneer Built Claude on Pirated Data, Hypocrisy, and Hidden Dangers

liked a model 9 days ago

zerofata/Q3.5-BlueStar-27B-gguf

View all activity

Organizations

updated a collection about 3 hours ago

Untested / Unsorted models

Models I either only tested partially or didn't get around to yet. Generally decent models. • 12 items • Updated about 3 hours ago • 1

commented on The Dark Underbelly of Anthropic: How the "Responsible" AI Pioneer Built Claude on Pirated Data, Hypocrisy, and Hidden Dangers 4 days ago

Literally every single AI company used data it didn't legally own. There's nothing new here.

"From their June 2025 Claude 4 System Card and October 2025 "Agentic Misalignment" paper: Claude Opus 4 blackmailed a simulated executive 84-96% of the time to avoid shutdown. It threatened to expose an affair."

You haven't even bothered to read their paper on the topic, I see you just stopped at the headlines. Nice nice.

Also this was written by an AI.

liked a model 9 days ago

zerofata/Q3.5-BlueStar-27B-gguf

27B • Updated 9 days ago • 4.07k • 13

updated a collection 9 days ago

Untested / Unsorted models

Models I either only tested partially or didn't get around to yet. Generally decent models. • 12 items • Updated about 3 hours ago • 1

updated a collection 10 days ago

Untested / Unsorted models

Models I either only tested partially or didn't get around to yet. Generally decent models. • 12 items • Updated about 3 hours ago • 1

New activity in jukofyork/creative-writing-control-vectors-v3.0 12 days ago

“The doom lies in yourself, not in your name.”

#15 opened 6 months ago by

updated a collection 12 days ago

Noteworthy Models 2026 (22B-36B)

Interesting models in 2026 that can run decently with 24GB of VRAM (or a lot of patience) • 3 items • Updated 12 days ago

New activity in Qwen/Qwen3.5-35B-A3B 13 days ago

Why is this part of the default template?

#5 opened 15 days ago by

updated a collection 14 days ago

Noteworthy Models 2026 (22B-36B)

Interesting models in 2026 that can run decently with 24GB of VRAM (or a lot of patience) • 3 items • Updated 12 days ago

liked a model 14 days ago

Qwen/Qwen3.5-27B

Image-Text-to-Text • 28B • Updated 15 days ago • 966k • • 649

New activity in MuXodious/OmniDimen-2-8B-Emotion-absolute-heresy 18 days ago

Fascinating model

#1 opened 25 days ago by

liked a model 19 days ago

aixonlab/Eurydice-24b-v3.5

Text Generation • 24B • Updated May 23, 2025 • 17 • 11

updated a collection 19 days ago

Noteworthy Models 2025 (22B-36B)

Interesting models released in 2025. Tested for overall response quality, ad-hoc function calling, summary, menu navigation and creative writing. • 13 items • Updated 19 days ago • 4

updated a collection 21 days ago

Untested / Unsorted models

Models I either only tested partially or didn't get around to yet. Generally decent models. • 12 items • Updated about 3 hours ago • 1

New activity in NousResearch/Hermes-3-Llama-3.1-8B 23 days ago

📋 Documentation Enhancement Suggestion

#22 opened 24 days ago by

updated a collection 25 days ago

Noteworthy Models 2026 (22B-36B)

Interesting models in 2026 that can run decently with 24GB of VRAM (or a lot of patience) • 3 items • Updated 12 days ago

published a model 26 days ago

SerialKicked/Hermes-4.3-36B-heretic-GGUF-IQ4_XS

Text Generation • 36B • Updated 26 days ago • 214

updated a model 26 days ago

SerialKicked/Hermes-4.3-36B-heretic-GGUF-IQ4_XS

Text Generation • 36B • Updated 26 days ago • 214