Tag
Explore by tags
A healed 64-layer 'frankenmerge' that stacks two Qwen3.5-derived finetunes into an ~18B GGUF model for multilingual text generation, reasoning, and reliable code/frontend output. Healed with a 1000-step QLoRA to reduce layer-boundary artifacts and targeted to run on 12–16 GB GPUs.
Provides 34k execution-style agent trajectories (11,766 issues) for supervised fine-tuning of code-focused LLMs. Each instance includes multi-step interactions, tool-call records, and final unified diffs; generated with Qwen3-Coder and released under permissive licenses for commercial use.
Runs goal-driven penetration tests by orchestration of an LLM agent and an MCP toolchain to perform reconnaissance, vulnerability discovery, exploitation, and structured PoC/report generation; supports multiple LLM providers and local MCP integrations; for authorized security testing only.
Provides a GGUF-packaged, native-INT4 quantized build of the multimodal Kimi K2.6 model for image-text-to-text inference — packaged for local/self-hosted inference engines (vLLM, SGLang, KTransformers) to reduce footprint while keeping multimodal capabilities.
Terminal-first developer workspace with an agentic AI side-panel that runs against your API keys or local models. Bundles a native PTY terminal, CodeMirror editor with AI edit diffs, file explorer, git history/graph, and a web preview in a ~7–8MB desktop app with no telemetry.
Provides a single MCP endpoint that lets AI coding agents search AWS docs, run sandboxed Python scripts, and make authenticated AWS API calls with enterprise guardrails like IAM condition keys, CloudWatch metrics, and CloudTrail auditing.
Collects real-world developer–AI coding sessions with full transcripts, tool calls, agent thinking traces, Git commits, and agent vs. human code attribution. Packaged as Parquet tables (conversations, sessions, commits, checkpoints, repositories) for analysis of agent behavior and human–AI collaboration.
Provides multiple GGUF-quantized exports of Carnice V2 (a merged BF16 SFT of Qwen3.6-27B) optimized for llama.cpp and Hermes-style agent traces, with quant tiers targeted at 16–24GB local GPUs and agentic inference.
Open-source Mixture-of-Experts LLM designed for extremely long-context (up to 1M tokens) text generation and agentic workflows; uses a hybrid attention + MTP design to reduce KV-cache footprint while enabling 42B active parameters and FP8 mixed-precision training.
Unified omnimodal foundation model for text, image, video and audio understanding and agentic workflows, with support for up to 1M-token context. Combines a sparse MoE LLM backbone, dedicated vision/audio encoders, multi-token prediction, and a hybrid sliding-window + global attention design to reduce KV-cache overhead.
Provides 1.7M agent interaction traces in terminus-2 format for training and evaluating agentic LLMs and RL agents. Compiled from 219 source datasets across code repair, shell, math, competitive programming and general tasks; produced with the Harbor harness.
Aggregates 750k+ Harbor-compatible agentic tasks from 100+ public sources (Parquet shards preserved). Includes tasks with and without verifiers for RL evaluation or SFT/datagen workflows, enabling reproducible trace generation.