Presents Skill Self-Play (Skill-SP), a co-evolutionary training loop where a proposer, solver, and dynamic skill controller generate, solve, and verify tasks conditioned on reusable skills — balancing verifiable execution with open-ended task diversity to boost LLM tool-use and reasoning.
AI Model2026
Delivers image and video understanding plus a built-in event‑gated streaming gate — a unified 4B multimodal foundation model that uses codec-aligned tokenization to cut visual tokens by >75% and yield up to 3.5× wall‑clock inference speedup for streaming and long‑horizon video tasks.
Transforms open-ended LLM optimization into self-verifiable reinforcement learning by turning tasks into proxy environments that produce deterministic, rule-based rewards. Proposes RLSVR and SpyRL — an information-asymmetric self-play scheme where agents vote to identify a preassigned spy, yielding verifiable rewards without human annotation. Demonstrated on summarization, creative writing and mathematical reasoning.
AI Dataset2026
Provides a 57,937-row, quality-filtered multi-teacher SFT distillation corpus combining outputs from Qwen3.8-Max, GLM-5.2 and Kimi K3 across math, code, reasoning, tool-use and dialogue. Includes 24 parquet training views (including a pre-tokenized GLM-4.7 view), configurable sampling weights (sft_balanced), and explicit tool-call trajectories for agent training.
AI Agent Papers2026
Bridges the proprietary-to-open-source gap in agentic search by converting multi-step retrieval and reasoning traces into a structured, style-normalized JSON protocol and using it for joint distillation + RL. Produces denser supervision that improves student success rates while reducing style drift.
AI Video2026
Generates synchronized stereo audio and video from multimodal inputs (text, images, video, audio), producing 4–15s clips at 24 FPS with a 768p base and an in‑context regeneration path to 2K; supports first/last‑frame and multi‑reference modes and ships as two task‑specific checkpoints.
Co-evolves a solver skill and a rubric-generator skill for text-space LLM optimization under decoupled objectives to avoid rubric gaming without using gold rubrics. Solver updates use criterion-level feedback; generator updates use independent audits of requirement coverage and response discrimination.
AI Agent Papers2026
Designs and evaluates a foundation GUI agent that performs cross-platform GUI and CLI actions on real devices to complete long-horizon workflows. Emphasizes a unified action space, a large-scale real-device mobile runtime, an AutoResearch-style data flywheel, and online RL training across 10,000+ concurrent environments.
AI Image2026
Provides a single-file ComfyUI-compatible LoRA plus full-weight RMSNorm/modulation .diff deltas that reproduce a Krea 2 fine-tune when applied to Krea 2 checkpoints. Rank-256 adapters, ~1.88 GB, MIT-licensed.
AI Model2026
Finetunes Qwen3.6‑35B using an adversarial generator–critic loop that synthesizes ~10,000 verifiable “frontier” tasks to boost scientific research, long‑horizon reasoning, coding and tool use; supports an extended 262,144‑token context and common serving stacks.
AI Model2026
A 35B additive ternary-quantized variant of Mach-1 that aims to preserve most capabilities of its BF16 teacher while reducing compute and memory; shows ~95% mean retention across 12 benchmarks and competitive per-task parity on several evaluations.
AI Model2026
Provides GGUF-quantized, ComfyUI-ready MiniMax‑H3 model files (FL2VA/REF2VA, text encoder, audio/video VAEs) to enable local ComfyUI inference for short video + stereo audio generation; requires the official VAEs and sufficient VRAM.