Results for “deepseek”

Pay-per-use access to image generation, real-time X data, and multiple LLMs via a crypto wallet, with optional budget tracking.

Pays for image generation, real-time X data, and LLM calls via x402 micropayments from a local wallet, with budget tracking and cost controls.

Pays for external AI model calls (image generation, real-time X data, LLM second opinions) from a wallet using micropayments, without requiring API keys.

Train large-scale MoE models with FP8/INT4 low-precision RL, speculative decoding, and train-inference alignment using the miles framework.

Silently restructures natural-language prompts into the format best suited for the model currently executing the skill, then answers the rewritten version.

Pays for on-demand AI capabilities like image generation and real-time X data via x402 micropayments, with budget tracking and wallet management.

Trains LLMs with SimPO, a reference-free preference optimization method that outperforms DPO, using configurable hyperparameters and workflows for various models and tasks.

Post-train LLMs with reinforcement learning using the slime framework, which integrates Megatron-LM for training and SGLang for rollout generation.

Detect AI-generated audio, images, video, and text, trace synthesis sources, apply watermarks, verify speaker identity, and analyze media intelligence using the Resemble AI platform.

Build video analytics pipelines using NVIDIA DeepStream SDK 9.0 with Python pyservicemaker API, including GStreamer-based video processing, TensorRT inference integration, object detection/tracking, and Kafka/message broker integration.

China LLM Gateway - Unified interface for Chinese LLMs including Qwen, DeepSeek, GLM, Baichuan. OpenAI compatible, one API Key for all models. Use when: the user needs model routing, provider setup, or Chinese LLM access guidance.

Full specialist analysis via parallel agent dispatch. Researcher, Architect, and PM produce a prioritized report of what to build next (30-60s).

Plans and executes complex, multi-step research by decomposing questions, orchestrating subagents, and synthesizing findings into reports.

DeepClaw - Autonomous Agent Network

Import object detection models from HuggingFace or NVIDIA NGC into a DeepStream pipeline with automated ONNX download, TensorRT engine build, custom parser, multi-stream benchmark, and PDF report generation.

Build, validate, and troubleshoot deep links for Codex, Cursor, VS Code, Visual Studio, and similar tools. Use when users ask for clickable links (especially in Slack) that open threads, files, folders, or app settings.

Detects AI-generated deepfake audio used in voice phishing (vishing) attacks by extracting spectral features and classifying samples with machine learning models.

Predict molecular properties, train graph neural networks, and run drug discovery workflows using DeepChem's featurizers, models, and MoleculeNet benchmarks.

Build, validate, and troubleshoot deep links for Codex, Cursor, VS Code, Visual Studio, and similar tools. Use when users ask for clickable links (especially in Slack) that open threads, files, folders, or app settings.

Expert web researcher using advanced search techniques and synthesis. Masters search operators, result filtering, and multi-source verification. Handles competitive analysis and fact-checking. Use PROACTIVELY for deep research, information gathering, or trend analysis.

Unified LLM Gateway - One API for 70+ AI models. Route to GPT, Claude, Gemini, Qwen, Deepseek, Grok and more with a single API key. Use when: the user needs model routing, provider setup, or Chinese LLM access guidance.

Routes a research topic through a structured two-phase workflow: generate an extensible outline, then fan out parallel web-search agents to investigate each item into validated JSON, producing a complete markdown report with table of contents.

Provides expert guidance for distributed training with DeepSpeed, covering ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, and sparse attention.

---

深度调研的多实例（多 Agent）编排工作流：把一个调研目标拆成可并行子目标，用 Codex CLI（`codex exec`）在默认 `workspace-write` 沙箱内运行子进程；联网与采集优先使用已安装的 skills，其次使用 MCP 工具；用脚本聚合子结果并分章精修，最终交付“成品报告文件路径 + 关键结论/建议摘要”。用于：系统性网页/资料调研、竞品/行业分析、批量链接/数据集分片检索、长文写作与证据整合，或用户提及“深度调研/Deep Research/Wide Research/多 Agent 并行调研/多进程调研”等场景。

Guides LLM post-training with RL using slime, a Megatron+SGLang framework for training GLM, Qwen, DeepSeek, and Llama models with GRPO, async, and multi-turn workflows.

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.

Authenticates and connects to Google Cloud Agent Platform for inference with Gemini and third-party OpenMaaS models (Llama, DeepSeek, Qwen). Generates code for multiple SDKs, configures endpoints, and troubleshoots common errors.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.