Loading...
Loading...
Found 45 Skills
Create basic Transformers model adapters for msModelSlim. Implements required interfaces and completes a four-step verification workflow: generate test model -> full fallback quantization -> weight verification -> quantization description validation. Use this skill when the user wants to: (1) create msModelSlim adapters for decoder-only LLM, (2) adapt understanding VLM text backbones for quantization, (3) implement W8A8/W4A16 quantization workflow for new models. Trigger: user mentions "msModelSlim", "adapter", "model adapter","quantization", "W8A8","W4A16", "transformers", "LLM", "VLM", "adapter creation", "适配器","模型适配", "量化", "模型适配器", "LLM量化"
Optimize an exported getitune model (the Geti training library) with post-training quantization. Use when a user wants to run `OVEngine.optimize()` / `engine.optimize()` to produce an INT8 model via NNCF, understands calibration-set requirements, or needs to re-validate and run inference with a quantized model versus the original FP32/FP16 model. Covers OpenVINO NNCF post-training quantization and the accuracy/size trade-off.
Diagnoses and improves Qdrant search relevance. Use when someone reports 'search results are bad', 'wrong results', 'low precision', 'low recall', 'irrelevant matches', 'missing expected results', or asks 'how to improve search quality?', 'which embedding model?', 'should I use hybrid search?', 'should I use reranking?'. Also use when search quality degrades after quantization, model change, or data growth.
Expert skill for using TileKernels, a library of optimized GPU kernels for LLM operations (MoE routing, quantization, transpose, engram gating, Manifold HyperConnection) built with TileLang.
Vector search best practices for Azure DocumentDB using `cosmosSearch` — choosing between DiskANN / HNSW / IVF, creating indexes, tuning `lBuild` / `lSearch` / `maxDegree`, Product Quantization (up to 16,000 dims), half-precision (fp16) indexing, and normalizing embeddings for cosine similarity. Use when building RAG / semantic-search applications, creating a vector index, tuning recall/latency, or reducing vector-index memory footprint.
Serve a model with MAX's `max serve` command: set up the environment (pixi or uv with the max-nightly conda channel / nightly wheel index), point the server at a Hugging Face repo or local checkpoint, target a custom architecture with `--custom-architectures`, and pick the right serve flags for the model. Use this whenever the user wants to run, launch, start, or host a model on MAX, bring up an OpenAI-compatible endpoint, serve a custom/ported architecture, debug a `max serve` startup failure, or figure out which serve flags (devices, quantization-encoding, max-length, task, trust-remote-code) a given model needs, even if they don't say "max serve" by name.
Develop, debug, and optimize SGLang LLM serving engine. Use when the user mentions SGLang, sglang, srt, sgl-kernel, LLM serving, model inference, KV cache, attention backend, FlashInfer, MLA, MoE routing, speculative decoding, disaggregated serving, TP/PP/EP, radix cache, continuous batching, chunked prefill, CUDA graph, model loading, quantization FP8/GPTQ/AWQ, JIT kernel, triton kernel SGLang, or asks about serving LLMs with SGLang.
Evaluates accuracy of quantized or unquantized LLMs using NeMo Evaluator Launcher (NEL). Triggers on "evaluate model", "benchmark accuracy", "run MMLU", "evaluate quantized model", "accuracy drop", "run nel". Handles deployment, config generation, and evaluation execution. Not for quantizing models (use ptq) or deploying/serving models (use deployment).
PR-backed and current-main optimization manual for the `MiniMaxAI/MiniMax-M2` series, including M2, M2.1, M2.5, M2.7, and M2.7-highspeed. Use when Codex needs to recover, extend, or audit MiniMax-specific optimizations, TP QK norm/all-reduce behavior, parser contracts, distributed runtime behavior, quantized loading, or backend-specific validation.