Discover the Best AI Resources

makecindy/cindy

Combines multiple agent harnesses, models, and tools in one workspace so agents can work across local projects and apps. Workspace, memory, skills, and tools persist when switching harnesses.

Ling-3.0-flash

A 124B hybrid-linear Mixture-of-Experts language model optimized for instruction following, long-context reasoning and agentic workflows, activating ~5.1B parameters per token. Key features include a 256K native context (extendable to 1M), alternating KDA/MLA attention layers, and vLLM/SGLang inference support.

LTX-2.5

Generates synchronized audiovisual output from text, image, or audio prompts — a diffusion-based multimodal model with componentized weights (video/audio VAEs, multilingual text encoder, distilled transformer) and ready integration with HuggingFace pipelines and ComfyUI.

Spotify Portal AI Plugins

Brings Spotify Portal workflows into coding agents for software catalog search, service briefings, diagnostics, and guarded actions. Its optional shunt plugin delegates large-file reading and boilerplate generation to AiKA worker models.

Instella-MoE-16B-A3B-Think

A sparsely activated Mixture-of-Experts (MoE) causal language model with 16B total parameters and 2.8B active parameters per token, released with end-to-end checkpoints and training recipe; trained on AMD Instinct GPUs and licensed for research use.

The Stack v3 (HuggingFaceCode/stack-v3-train)

Provides a near-deduplicated, quality-filtered 15.9 TB training subset of GitHub source code grouped by repository, with inline UTF‑8 file contents and repo metadata for pre-training and analysis of code LLMs; cutoff Aug 7, 2025, ODC-By license.

AREX: Towards a Recursively Self-Improving Agent for Deep Research

Alternates targeted research and constraint-wise audits to recursively improve long-horizon answers: an inner loop gathers evidence and drafts solutions, an outer loop audits unresolved claims and launches focused follow-ups. Trains 4B dense and 122B-A10B MoE agents with long-horizon RL and agentic mid-training, outperforming comparable-scale baselines on multi-step research benchmarks.

Show, Don't Tell: Evaluating Spatial Cognition in Generative Pixels Rather Than LLM Text

Evaluates spatial cognition of image-generation models by eliciting protocol-constrained visual answers and parsing pixel outputs into structured predictions compatible with existing metrics. Introduces the ProVisE framework and SpatialGen-Bench (470 samples) to compare image-generation models and text-output VLMs on unified spatial tasks.

Visual Contrastive Self-Distillation

Converts image-content removal into a contrastive on-policy self-distillation signal: the EMA teacher produces next-token distributions with and without image content, uses their log-probability differences to sharpen visual-grounded candidates, and distills that full-distribution target into the student—no external teacher or extra inference cost.

KAT-Coder-V2.5-Dev

A text-only open-weight MOE code model (35B total, 3B active) fine-tuned with SFT+RL for agentic coding; achieves strong agentic-code benchmarks, supports 262k context and deployment via Transformers/vLLM; vision weights are not included.

PerceptionBench

Evaluates atomic visual perception of multimodal LLMs using 3,000 short visual questions that isolate ten perceptual skills. Built from an error taxonomy across 42 benchmarks, capability-balanced and accompanied by a model leaderboard.

Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills

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.