Builds an editable, persistent 3D world state to drive iterative previsualization for film, games, and design — enabling local edits and recombinations instead of one-shot video regeneration. Uses separate stages for state construction, state evolution, and state access, with render-feedback camera refinement.
Provides an FP8-post-trained 27B multimodal causal language model with a native vision encoder, large-context support (262,144 native, extensible to 1,000,000), controllable thinking-mode reasoning, and compatibility with common inference engines for deployment.
Supports multimodal scientific understanding, long-horizon agentic workflows and scientific tool interaction using a unified pipeline of multimodal pretraining, supervised fine-tuning and scalable multi-task reinforcement learning. Distinctive features include time-series modules for signal forecasting and a separate Memory Decoder that enables rapid domain specialization without changing the frozen 397B backbone.
Provides a 27B Qwen3.8 GGUF build for local/offline deployment, optimized with Unsloth Dynamic V3.0 quantization. Offers switchable thinking-mode, native vision-language understanding, and native long-context support (262k+ tokens).
A 27B Qwen3.8 vision‑language causal transformer quantized to NVFP4 for lower‑memory inference. Provides 262K native context (extensible to 1M), Unsloth Dynamic V3.0 4‑bit quantization and MTP support so Qwen3.8‑class multimodal workloads can run on 24GB‑class GPUs.
Turns natural-language instructions into end-to-end Android automation that drives real devices and captures Logcat and screenshots. Uses a multimodal “Dynamic‑First, Coordinate‑Fallback” locator and a native MCP server for IDE integration, supporting fast Flash (3–5s/step) and verifiable Pro workflows.
Conducts end-to-end multidisciplinary research directly from heterogeneous raw evidence using lifecycle-wide perception and three autonomous agents (Ideation, Experiment, Writeup). Integrates perceptual analysis, execution provenance, and code-enforced checks to produce executable analyses, validated results, and compiled manuscripts across many modalities.
Performs unified parsing of digital and camera-captured documents (layout, text, tables, formulas) using a ~1.2B-parameter vision–language model. Key differences: geometry-aware modeling, curvature-guided sampling, and content-structure decoupled training to handle real-world deformations without separate dewarping.
Provides 617.5 hours of high-precision optical motion-capture with synchronized object trajectories and standardized 55-joint BVH for whole-body and human–object interaction research. Frame‑LU indexed and paired with natural-language descriptions; designed for humanoid learning, motion priors, and interaction-aware benchmarks.
Parses digital and camera-captured documents into structured outputs (text, layout, tables, formulas, figures) using a lightweight (~1.2B) open-source vision-language model. Uses geometry-aware modeling, multi-node consensus pseudo-labeling, and content-structure decoupling to handle warped, photographed, and digital pages.
Provides locally runnable GGUF quantizations of Qwen3.8-27B with the MTP speculative-draft head preserved and a Heretic weight edit that substantially reduces refusal rate. Ships multiple quant sizes with published imatrix and perplexity measurements for local inference under Apache‑2.0.
Systematically evaluates AI-generated video detectors and generators for real-world crisis scenarios using RA-Bench (17,886 clips: 1,830 real anchors, 16,056 generated). Shows detector families fail to generalize across generation conditions, and that human-misleading videos and social dissemination further degrade detection.