AI Model2026
A distilled 26M-parameter encoder–decoder LLM for on-device function-calling and tool use. Uses a pure-attention Simple Attention Network, provides open weights and local finetuning, and targets high-throughput inference on the Cactus runtime.
AI Client2026
Orchestrates parallel CLI-based AI agents in isolated git worktrees so you can run multiple coding agents side-by-side, review AI-generated diffs, and link PRs/CI to each worktree. Desktop client with a mobile companion and BYO model subscriptions.
Hands-on, phase-based curriculum for building end-to-end AI systems from first principles — implement algorithms, run tests, and ship reusable artifacts (prompts, skills, agents, MCP servers) across Python, TypeScript, Rust, and Julia under an MIT license.
AI Dataset2026
A 228,557-example dataset of reasoning traces segmented into blocks with iterative, compressed "memento" summaries so LLMs can learn to manage long context. Includes a training-ready subset and a `full` subset with sentence/block-level annotations for research and SFT.
AI Agent2026
Scans AI agent skills for security issues—detecting vulnerabilities, malicious patterns, and supply-chain risks before installation. Combines static AST checks (64 patterns across 16 categories) with optional LLM semantic review, OSV live CVE lookups, and JSON/Markdown/SARIF outputs for CI or manual review.
AI API2026
Provides short-lived, copy-pasteable API tokens that let developers access 90+ LLMs (GPT‑5.5, Claude, Gemini, Grok, etc.) without a credit card or registration. Keys are refreshed multiple times daily, each carries a $20–$100 budget and expires in 24–48 hours. Works with any OpenAI-compatible client via a single base URL.
AI Dataset2026
Large-scale mid-training corpora for multimodal models: 10,809 ~60s video shards, caption splits (30s/60s/180s/>10min), 84 spatial-reasoning shards, and CSV mappings to source YouTube IDs. Small Parquet preview configs are provided for schema inspection.
AI Agent2026
Runs an autonomous self-improvement loop where a meta agent crafts a task-specific agent, a target agent executes trials, and a feedback agent updates both harness (code) and model weights—provider-agnostic profiles with reproducible runs and a live dashboard.
AI Dataset2026
Provides 1,000,000 model-generated chain-of-thought traces and instruction–response pairs for fine-tuning and distilled supervision. Focused splits (coding, PHD-Science, General-Math, MultilingualSTEM), ~5B tokens, Apache-2.0 license.
AI Model2026
Runs the Bonsai family of quantized LLMs locally (including vision-capable 27B): provides scripts and demo UIs to run 1-bit and ternary Bonsai models on macOS (Metal), Linux/Windows (CUDA/Vulkan/ROCm), or CPU, with long context, tool-calling and an optional Open WebUI agent demo.
AI Dataset2026
Provides multi-turn agent trajectories with real tool executions and explicit <think> reasoning blocks for training and evaluating tool-calling agents. Contains two model-sourced configs (Kimi-K2.5, GLM-5.1) totaling ~14.7K samples — useful for SFT, agent-skill research, and tool-integration experiments.
AI Video Papers2026
Provides a diagnostic suite that audits video-understanding benchmarks to find samples solvable without visual or temporal input, filters those shortcuts, and produces a distilled video-native testbed that reveals major capability gaps in current Video-LLMs.