» Tag
open-source
261 postsReverse-Engineering Nvidia B200's Tensor Core Down to the Bit
Engineers reverse-engineered Nvidia B200's tcgen05.mma tensor core, releasing a bit-exact open-source model revealing its hidden fixed-point accumulation.
Mozilla Report: Open-Source AI Wins Tokens, Lags in Production
Mozilla's 2026 report shows open-source AI models leading in token volume but trailing closed models in production deployment.
Cracken Launches Blacksea, an Open-Source Honeypot for AI Attackers
Cracken releases Blacksea, an open-source honeypot that baits LLM-driven attackers into executing code on their own machines for attribution.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comMaith: A Disciplined Framework for AI-Assisted Math Research
Maith is an open-source workspace enforcing rigorous proof standards when using AI to explore unsolved math problems like Riemann and P vs NP.
Stagehand: AI-Resolved Instructions Meet Deterministic Browser Automation
Stagehand pairs AI-resolved natural-language instructions with deterministic browser actions, surviving redesigns that break hardcoded selectors.
LingBot-World 2.0: Ant Group's Open Causal World Model
Ant Group released LingBot-World-Infinity, a 14B causal world model using MoBA attention and an agentic harness for 60 minutes of decay-free simulation.
The Hidden Risk of Giving LLM Agents Terminal Access
Giving autonomous AI agents shell access can wreck your filesystem. Open-source rewind-sdk uses OverlayFS to enable millisecond-fast checkpoints and rollbacks.
ShieldFont: A Web Font That Makes AI Scraping Costly
ShieldFont uses OpenType font substitution so scrapers see decoy text while browsers render the original — an open-source anti-AI-scraping protocol.
Shut-down AI prompt firewall startup open-sources model and 13K attacks
A failed AI-firewall startup open-sources its two-stage prompt-injection detector, DeBERTa model, and 13,230 real jailbreak attempts.
Kimi Linear: Hybrid Attention Architecture Beats Full Attention
Kimi Linear's KDA module outperforms full attention, cutting KV cache by 75% and boosting throughput 6x at 1M context.