Transfers RL-induced policy shifts from a smaller 'weak' teacher to a stronger target by using the teacher's post-/pre-RL log-ratio as a dense implicit reward applied on the student's on-policy states. Enables reuse of RL supervision without running RL rollouts on the target, improving sample/time efficiency.
AI Model2026
GGUF-format quantized release of DeepSeek‑V4‑Flash for local inference — compatible with llama.cpp and Unsloth runtimes, with guidance for FP4/FP8 mixed precision and Q4/Q8 quantization; tuned for million-token long-context usage.
AI Model2026
Provides pre-converted colibrì-format int4 weights so GLM-5.2 (744B MoE) can run by streaming routed experts from disk on a consumer machine with ~25 GB RAM. Includes MTP shard for lossless speculative decoding; requires the colibrì engine and ~400 GB NVMe.
AI Model2026
27B multimodal LLM post-trained to prioritize agentic, weight-scaled reasoning over 64K-token contexts. Built on Qwen3.6-27B and released with BF16 weights plus several GGUF quants; aimed at coding, long-document reasoning, tool use and multimodal inspection.
Expresses diverse computer-vision tasks as instruction-driven text, image, or mixed generation from a single unified multimodal model, producing outputs for detection, segmentation, depth, pose, OCR and more. Trained on a converted SenseNova‑Vision instruction–response corpus and requires no task-specific prediction heads.
AI Dataset2026
Provides ~5M model-generated reasoning chains (within 5k sequence length) with structured fields for supervised fine-tuning, reasoning distillation, and instruction tuning. Includes separate fields for prompt, reasoning trace, final answer and a ChatML view; streaming access recommended for large-scale use.
Creates an open-ended interactive world simulator with an unbounded interaction horizon via causal pretraining, a distilled real-time runtime that drives 720p@60fps, a wider action/event repertoire, and a pilot–director agent split for behavior planning and environment synthesis.
AI Dataset2026
Provides ~5M tokens of chain-of-thought reasoning traces generated by many LLMs (DeepSeek, Qwen, Gemma, etc.) for training and evaluating reasoning SLMs — includes repo_id, tok_len, user, thought_trace, assistant and ChatML fields; sequences limited to 5k.
Provides IdeaGene-Bench, a dataset and evaluation suite for scientific-lineage reasoning and lineage-grounded idea generation, representing papers as minimal, typed Idea Genome objects and GenomeDiffs that record inheritance, mutation, loss, import and novel insertion. Includes 1,961 lineage traces, IG-Exam (42 task types) and IG-Arena with a Population-Evolution Score for generation.
AI Agent Papers2026
Provides a terminal-style benchmark of 46 long-horizon tasks decomposed into fine-grained graded subtasks to produce dense intermediate rewards and partial credit, enabling evaluation of long-horizon planning, long-context management, and iterative debugging. Tasks typically require hundreds of episodes and minutes-to-hours of execution; baseline evaluations report high token and episode consumption with low pass rates, highlighting evaluation headroom.
AI Agent Papers2026
Evaluates proactive, multimodal agents on 400 bilingual real‑world tasks across five capability axes (Skill Usage, Exploration, Long‑Context Reasoning, Multimodal Understanding, Cross‑Platform Coordination) using live Docker‑based, stepwise closed‑loop evaluation to separate base model skills from framework design.
AI Model2026
Provides GGUF-quantized builds of the Qwythos-9B-v2 LLM for local runtimes, with multiple quant levels, optional MTP-enabled variants, a 1,048,576-token context window, and an optional BF16 vision projector for multimodal use.