DEV Community

FX-LgLL profile picture

FX-LgLL

404 bio not found

Joined Joined on 
THX-01 Just Processed 70 Million Tokens. Our Open-Source Decision Model Is Just Getting Started.

THX-01 Just Processed 70 Million Tokens. Our Open-Source Decision Model Is Just Getting Started.

Comments
3 min read
We Open-Sourced a 322M Decision Model. Now Its API Is Free

We Open-Sourced a 322M Decision Model. Now Its API Is Free

Comments 1
3 min read
We made our decision model's API free (and the weights are open)

We made our decision model's API free (and the weights are open)

1
Comments
2 min read
We Open-Sourced THX-01: A 322M AI Decision Model Small Enough to Run on a Smartphone

We Open-Sourced THX-01: A 322M AI Decision Model Small Enough to Run on a Smartphone

Comments
2 min read
We open-sourced THX-01: a 322M decision model that matches Sonnet 5.5 on ticket classification at ~10 ms (Apache 2.0, pip install thx01)

We open-sourced THX-01: a 322M decision model that matches Sonnet 5.5 on ticket classification at ~10 ms (Apache 2.0, pip install thx01)

Comments
2 min read
SIMURG: A Guard That Catches LLM Hallucinations Mid-Stream and Heals Them

SIMURG: A Guard That Catches LLM Hallucinations Mid-Stream and Heals Them

1
Comments 2
4 min read
Free Unlimited Web Search for AI Agents

Free Unlimited Web Search for AI Agents

Comments
1 min read
I found a completely free web search for AI agents.

I found a completely free web search for AI agents.

Comments
1 min read
I almost killed all GLM 5.3 Flash 2 bit hallucinations, with SIMURG

I almost killed all GLM 5.3 Flash 2 bit hallucinations, with SIMURG

1
Comments
2 min read
GLM-5.3-Flash is a total blast when you kill its hallucinations

GLM-5.3-Flash is a total blast when you kill its hallucinations

Comments
1 min read
I just killed hallucinations on my 2 bit Qwen3.8 27B

I just killed hallucinations on my 2 bit Qwen3.8 27B

Comments
1 min read
Five detectors, one bad merge: why our LLM corruption guard flagged 43% of healthy output

Five detectors, one bad merge: why our LLM corruption guard flagged 43% of healthy output

Comments 2
6 min read
loading...