Qwen · For developers

Run a one-shot or flat parallel batch of provider LLM subagents (headless cc-fleet subagent) that return a result.

ethanhq/cc-fleet

Guided workflow for adding a new model architecture to llama.cpp.

JakeATX/llamAmpere

Analyze the token length of an OT-Agent conversation-format (ShareGPT-style) dataset — the per-trace distribution (median/p90/max) and/or counts under a token threshold + a metadata predicate (e.g.

open-thoughts/OpenThoughts-Agent

Hardware-aware adaptive Hermes runtime using portable Turbofiles, total usable memory, owned native llama.cpp residency, stable auto/active:main/active:aux routes, and evidence-backed promotion.

SouthpawIN/turbofit

Serves AI models on AMD Instinct GPU hardware using vLLM. An agent skill from amd/skills.

amd/skills

Measures the throughput difference between two branches with force-balanced routing.

mlc-ai/pith-train

企业微信 飞书 钉钉 淘宝 小红书 抖音电商 微信小程序 微信小店 拼多多 有赞 微信支付 支付宝 京东 SHEIN 得物 火山引擎 阿里云百炼 泛微 北森 API文档搜索。

LiangNiang/OpenMantis

Draw Sebastian-Raschka-gallery-style TikZ architecture diagrams for any HuggingFace decoder-only LLM, with per-block parameter formulas and concrete numbers.

yzlnew/infra-skills

A skill your agent uses when creating complete generated audio with qwen-audio-3.1-tts-next, including podcasts, radio drama, advertisements, multiple speakers, reference voices, ambience, sound…

tjxj/z-skills

Verify a built wally binary against a pinned C++ desktop kit on macOS and Windows.

RunanywhereAI/wally

Run HomeRail's built-in read-only pull request review DAG. An agent skill from xiaotianfotos/homerail.

xiaotianfotos/homerail

Call this skill when you need to debug CUDA crashes in SGLang using kernel API logging

sgl-project/sglang

Generate an e2e profiling trace of an SGLang server run. An agent skill from sgl-project/sglang.

sgl-project/sglang

Generate or edit images via OpenRouter Image API across top models (Nano Banana, GPT Image, FLUX.2, Seedream, Qwen, Grok).

rtadewald/skills

Guide a tester through hipfire bring-up, serve smoke, claim-scoped harnesses, and benchmark reporting on AMD RDNA/CDNA GPUs.

warpfront/hipfire

Has several open-weight models answer a question, rank each other's anonymized answers, then lets a chairman model write the final response through Fireworks AI.

dair-ai/dair-academy-plugins

Builds and publishes a Gradio demo on Hugging Face Spaces for a LoRA, with the pipeline, UI and settings chosen to match that LoRA's task and model card.

huggingface/skills

Estimate GPU memory usage for Megatron-based MoE (Mixture of Experts) and dense models.

yzlnew/infra-skills

Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral).

Orchestra-Research/AI-Research-SKILLs

Provides guidance for enterprise-grade RL training using miles, a production-ready fork of slime.

Orchestra-Research/AI-Research-SKILLs

播客/小宇宙 → 下载 → 转录 → 存为 Markdown 的完整工作流. An agent skill from chubbyguan/chubbyskills.

chubbyguan/chubbyskills

Generate text, have conversations, write code, reason, and call functions with Qwen models.

QianWen-AI/qianwen-ai

Review llama.cpp changes against project conventions and common reviewer pitfalls before a PR.

JakeATX/llamAmpere

Narration, sound effects and music beds in the voice the studio set up (ElevenLabs, Qwen3-TTS on this machine's GPU, or the owner's own voice from the recording booth): lines from the approved…

GTKottman/mortiflix-oss

Улучшить/оптимизировать/переписать промпт под ТЕКУЩУЮ модель по официальным правилам вендора (Claude Code, Codex, Gemini CLI, Grok, Kimi, Qwen) и объяснить почему.

edvardgrishin27/nativeprompt

Add new AI models to Kiln's mlmodellist.py and produce a Discord announcement.

Kiln-AI/Kiln

Control local desktop applications through Computer Use for tasks that require reading or operating app UI.

QwenLM/qwen-code

编写并运行 dynamic workflow：用 Claude Code 的 workflow 方言写一段简短的 JavaScript 脚本，再用 odw CLI 在宿主 agent 的上下文之外，把子任务扇出给 coding-agent CLI （Codex、Claude Code、Gemini、Qwen、Kimi 或自定义），后台跑完后只取回最终结果。

xz1220/open-dynamic-workflows

MUST be considered at task start in this repository to detect OpenSpec change lifecycle work and route Claude Code/CCR coordination tasks without loading unnecessary context.

HsienW/chat-gun

Breaks LLM torch profiler traces down by forward pass, layer and kernel, with timing tables and Perfetto time ranges for the layers you want to inspect.

BBuf/AI-Infra-Auto-Driven-SKILLS

DashScope (Alibaba Cloud Bailian / 阿里云百炼) integration — image generation (qwen-image-2.0-pro), text-to-speech (qwen3-tts-flash), and ASR with word-level timestamps (qwen3-asr-flash-filetrans).

calesthio/OpenMontage

Calculate training costs for Tinker fine-tuning jobs. An agent skill from sundial-org/skills.

sundial-org/skills

Set up the minimal set of artifacts (tokenized DCLM corpus shard + released HuggingFace checkpoint converted to DCP) required to benchmark, profile, or regression-test a MoE model in PithTrain.

mlc-ai/pith-train

Cloud GPU processing via RunPod serverless. An agent skill from digitalsamba/claude-code-video-toolkit.

digitalsamba/claude-code-video-toolkit

Run Agmente iOS end-to-end tests against a local ACP agent (Gemini, Claude, Qwen, or Vibe), validate core RPC flow, and perform mandatory cleanup.

rebornix/Agmente

Runs Qwen Code headlessly with JSON output for repository work, using restricted approvals and bounded turn limits, and never the --yolo flag.

CherryHQ/cherry-studio

Performance optimization guidelines for Splitrail. An agent skill from Piebald-AI/splitrail.

Piebald-AI/splitrail

LangChain / LangGraph engineering pitfalls and verified fixes.

ob-labs/agentseek

Create or update reusable Dynamic Workflow JavaScript files under .qwen/workflows.

QwenLM/qwen-code

A skill your agent uses when a new Ollama Cloud model is announced or available (e.g.

heypinchy/pinchy

Bilingual guide for merging ViT + LLM into LlavaOnevision2 HF checkpoint and validating weight/inference consistency

EvolvingLMMs-Lab/LLaVA-OneVision-2

Run a 4-hour multi-node Hyperloom Qwen3-30B-A3B optimization (Infera PD-disaggregated or RayJob aggregated) with --nodes 2 and sglang MoE tuning on MI325X.

AMD-AGI/Hyperloom

Audit SGLang startup logs, save evidence, and propose cleanup for user review.

sgl-project/sglang

MUST be considered at task start in this repository to detect OpenSpec change lifecycle work and route Qwen Code review tasks without loading unnecessary context.

HsienW/chat-gun

Orchestrate the implement-review loop for Spec Kitty work packages using any configured agent.

spec-kitty/spec-kitty

给 FlashRec 引擎接入一个新模型架构（新的 HF checkpoint / 非 Qwen3 结构）。涵盖模型定义、权重合并加载、FP8 双路径、融合 kernel 接线、CUDA graph 兼容、精度校验、以及压测+trace 验证闭环。当用户要"增加/支持/接入新模型"时使用。

sohu-mptc/FlashRec

Hands coding work such as codebase questions, bug fixes, refactors and PR reviews to the Qwen Code CLI, driven through acpx instead of scraped terminal sessions.

QwenLM/qwen-code

Diagnoses other Qwen Code sessions on your machine that look frozen or slow, checking process state, CPU, memory, child processes and logs, and reports what it finds.

QwenLM/qwen-code