Results for “ray-dalio”

Scales Python ML workloads across clusters using Ray's distributed tasks, actors, data, and serving capabilities.

Train large language models (7B-70B+) with RLHF using PPO, GRPO, DPO, and other algorithms, accelerated by Ray and vLLM for distributed multi-GPU setups.

Documents the Raycast design system with color tokens, typography hierarchy, and surface modes for building consistent dark-mode interfaces.

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.

Write, review, and migrate code using NVIDIA DALI's imperative dynamic-mode API for efficient data loading and preprocessing.

DALL-E: Zero-Shot Text-to-Image Generation

ALIAS for dag-orchestrator. The original DAG execution skill, now unified with orchestrator into dag-orchestrator. Use dag-orchestrator for the full HTDAG experience.

Dao reference tool. Use when working with dao in blockchain contexts.

Adversarial reviewer that stress-tests claims and conclusions. Use when the user asks to talk to KAY/O, requests an adversarial review, or needs claims validated before publication.

COCO: Microsoft COCO: Common Objects in Context

Turn "is this deal real?" into a deal-by-deal risk verdict. An 8-state risk taxonomy, a qualification-vs-stage gap check, multi-thread coverage, days-dark per deal, and a commit-creep watch. Built for B2B sales teams, customizable to your CRM and your sales process. Trigger on "how are my deals", "which deals are stuck", "what's at risk this month", "is the rep sandbagging", "multi-thread check", or any open-deal diagnostic.

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.

RHINO — Momentum pyramider. Top 10 assets by OI + volume. Enters small (30% of max) on high-conviction convergence, then adds to winners at +10% ROE (40% more) and +20% ROE (final 30%). Thesis re-validated before every add — 4h trend intact, SM aligned, volume present. DSL High Water Mode trails the full position. The only skill in the zoo that builds into winners instead of entering full size and hoping.