All categories
DevOps & Cloud
Deployment, cloud, containers, CI/CD, monitoring, and infrastructure
- 20,326
- 847
Security grades appear on each card once the skill has been scanned. Newly imported skills may briefly show without a grade until the backfill job runs.
Open in full browserBrowse devops & cloud skills
Showing 9,337–9,360 of 20,326 skills
- Mix Reasoning Switching ModesImprove reasoning efficiency by dynamically adjusting elaboration depth within a single response. Difficult steps receive extensive chain-of-thought; straightforward steps use concise inference. Mix reasoning modes to achieve 57-79% token savings under compute budgets while maintaining accuracy on math, STEM, and coding benchmarks.Votes: 0GitHub stars: 6
- Mirror Speculative DecodingRun draft and target models in parallel across GPU/NPU using bidirectional speculation: draft predicts forward continuations while target speculates correction paths simultaneously. Achieve 2.8x-5.8x wall-time speedup on 14B-66B models.Votes: 0GitHub stars: 6
- Minicpm4 Efficient LlmsBuild ultra-efficient language models for edge devices using sparse attention, high-quality data filtering, and ternary quantization, achieving Qwen3-8B performance with 22% of training tokens.Votes: 0GitHub stars: 6
- Metaclaw Continual Agent LearningEnable LLM agents to evolve behavioral skills and policies online through skill synthesis from failures and opportunistic gradient-based refinement, without service interruption.Votes: 0GitHub stars: 6
- Lspo Length Aware Dynamic SamplingFilter training samples by response length to identify high-confidence correct solutions and complex problems, improving sample efficiency in LLM reasoning RL without additional metrics.Votes: 0GitHub stars: 6
- Longvila Scaling Rl Long VideosTrain vision-language models on hour-long videos using a two-stage pipeline combining supervised fine-tuning with GRPO, reaching 71% accuracy on VideoMME while supporting 8,192 frames through efficient multi-modal sequence parallelism.Votes: 0GitHub stars: 6
- Jit Spatial Diffusion AccelerationAccelerate diffusion transformers by processing only sparse anchor tokens in early stages, then expanding to full spatial resolution using learnable extrapolation. Combines SAG-ODE velocity field lifting with importance-guided token activation for lossless speedup.Votes: 0GitHub stars: 6
- Janus Moe DisaggregationEnable scalable MoE inference by disaggregating attention and expert layers onto independent GPU sub-clusters. Use adaptive two-phase communication, activation load-balanced scheduling, and activation-aware expert management. Achieve 3.9× higher per-GPU throughput than state-of-the-art systems.Votes: 0GitHub stars: 6
- Heteroscale AutoscalingScale disaggregated LLM inference (prefill-decode) via topology-aware scheduling and metric-driven policies, achieving 26.6% GPU utilization improvement and conserving hundreds of thousands GPU-hours dailyVotes: 0GitHub stars: 6
- Growing Transformers Modular ExpansionBuild language models layer-by-layer on frozen embeddings, training new Transformer blocks sequentially while keeping lower layers frozen. Achieves 5% improvement over monolithic baselines on MMLU while fitting 740M trainable parameters per stage on single GPUs, enabling resource-efficient incremental scaling to 2.3B parameters.Votes: 0GitHub stars: 6
- Gem Agentic Llm EnvironmentA standardized environment framework for training and evaluating LLM agents, providing 24+ tasks with asynchronous vectorized execution, extensible wrappers, and integration examples for five RL frameworks. Enables reproducible agent research and training at scale.Votes: 0GitHub stars: 6
- Flowrl Reward Distribution MatchingTrain LLMs with distribution-matching rewards instead of reward maximization to achieve 10% improvement on math reasoning while improving solution diversity by matching the full reward distribution via flow balancing, addressing mode collapse in long chain-of-thought tasks.Votes: 0GitHub stars: 6
- Fairy2i Complex Llm QuantizationEnable extreme 1-2bit quantization of pre-trained LLMs by leveraging complex-valued arithmetic. Convert real-valued linear layers to complex domain losslessly, quantize to fourth roots of unity {±1, ±i}, and apply recursive residual error quantization for near full-precision performance.Votes: 0GitHub stars: 6
- F1 Vla Understanding Generation ActionsF1 is a 4.2B-parameter Vision-Language-Action model that reformulates robotic control as foresight-guided inverse dynamics. Rather than reactive state-to-action mappings, F1 predicts plausible future visual states and derives actions to achieve them, enabling robust planning in dynamic environments with 82.2% real-world task success rate.Votes: 0GitHub stars: 6
- Exevrm Video Reward ModelingBuild robust reward models for computer-use agents by processing execution videos with spatiotemporal token pruning (STP+TTP) to remove redundancy while preserving task-relevant UI details. Achieve 84.7% accuracy with adversarial instruction translation.Votes: 0GitHub stars: 6
- Eager Entropy ScalingMonitor token-wise entropy to adaptively allocate compute during inference. Branch into multiple paths at high-entropy tokens, reducing token generation by up to 65% while improving accuracy by up to 37% on reasoning tasks.Votes: 0GitHub stars: 6
- Docdancer Document AgentBuild open-source agents for document question-answering by modeling DocQA as information-seeking with explicit tool utilization. DocDancer uses an exploration-then-synthesis pipeline to generate high-quality training data, addressing the scarcity that limits agent-based document understanding systems.Votes: 0GitHub stars: 6
- Dirl Diffusion RlEnable effective RL for diffusion language models via DiPO (unbiased GRPO for dLLMs) and framework optimizations. FlexAttention accelerates blockwise training, LMDeploy optimizes inference, achieving training-inference consistency—improving dLLM math performance to rival larger autoregressive models.Votes: 0GitHub stars: 6
- Deep Search Mcts Rlvr TrainingOvercome exploration bottlenecks in reasoning RL by integrating Monte Carlo Tree Search during training (not just inference). Global frontier selection and entropy-guided sampling reduce GPU hours by 5.7x while improving performance.Votes: 0GitHub stars: 6
- Codev Verilog ReasoningGenerate Verilog hardware code from natural language using reasoning-enhanced LLMs, combining rule-based testbench generation with round-trip data synthesis and adaptive DAPO reinforcement learning for reliable hardware design.Votes: 0GitHub stars: 6
- Clipping Free Policy OptimizationReplace hard clipping in policy gradients with smooth quadratic penalties derived from Total Variation divergence constraints. Eliminates zero-gradient regions and training instability while maintaining stable policy evolution.Votes: 0GitHub stars: 6
- Batch Speculative DecodingFixes batch speculative decoding ragged tensor problem where sequences in batches accept different token counts, desynchronizing state. EQSPEC guarantees output equivalence through proper synchronization. EXSPEC reduces overhead 40% via cross-batch scheduling. Enables efficient parallel decoding with 95% equivalence.Votes: 0GitHub stars: 6
- Attention Residuals For Depth ScalingReplace uniform residual accumulation with depth-wise attention that selectively aggregates earlier layer representations. Improve gradient flow and model performance in deep architectures by learning content-dependent depth-wise selection.Votes: 0GitHub stars: 6
- Areal Async Rl Language ReasoningScale RL training to large models through decoupled rollout and training workers with controlled data staleness.Votes: 0GitHub stars: 6