profile-vllm-performance SKILL.md skillA
profile-vllm-performance from nvidia-ai-blueprints/video-search-and-summarization is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add nvidia-ai-blueprints/video-search-and-summarization/profile-vllm-performance@v3.3.0-rc0Pins v3.3.0-rc0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- nvidia-ai-blueprints/video-search-and-summarization
- Path
skills/benchmarking/profile-vllm-performance/SKILL.md- Versions recorded
- 1
- Latest version
- v3.3.0-rc0, committed Oct 1, 2026
- Size
- 6,102 bytes
- Repository stars
- 1,968
- License
- Apache-2.0
Profile vLLM and RT-VLM inference to identify and remove GPU-idle gaps, underfilled batches, transfer stalls, serialized multimodal work, scheduler gaps, or KV pressure. Use this skill when GPU utilization is unexpectedly low or bursty, capacity trails another runtime, TTFT or ITL regresses, or multimodal prefill and decode appear serialized. Not for running a first benchmark without a reproducible fixed-load signal.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Profile vLLM Performance Find the dominant loss mechanism, prove it with a correlated timeline, make the smallest causal change, and validate the result with a matched A/B run. A capacity number or average GPU-utilization sample is not bubble attribution. ## Freeze The Comparison Record the exact code, container, model revision, precision, hardware, scheduler settings, workload shape, media, prompt, token budget, sampling, cache state, concurrency, and success criteria. Prove semantic correctness before optimizing. Do not compare runtimes when any of these dimensions differ unless that dimension is the declared variable. ## Establish The Signal 1. Verify actual output-token counts, successful work, fresh-stream coverage, dropped chunks, cancellations, and cleanup. Early EOS or silent failure is not throughput. 2. Reproduce at fixed load below the knee, near the highest stable point, and just beyond it when practical. Keep capacity search separate from profiling. 3. Record TTFT, inter-token latency, throughput, queue delay, batch size, scheduled tokens, active sequences, preemptions, KV occupancy, GPU memory, and phase …
Read the whole file at its exact version.
How to install
mdr add nvidia-ai-blueprints/video-search-and-summarization/profile-vllm-performance@sha256:f0aae01465e8fcbbThe line at the top of this page pins the label v3.3.0-rc0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_h2dimp6hq374blt3)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 1, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (6102 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
nvidia-ai-blueprints/video-search-and-summarization · 1,968 stars · license Apache-2.0 · pushed 2026-11-07 · branch develop
API
GET https://markdownregistry.com/api/v1/artifacts/art_h2dimp6hq374blt3 GET https://markdownregistry.com/api/v1/resolve?ref=nvidia-ai-blueprints/video-search-and-summarization/profile-vllm-performance GET https://markdownregistry.com/api/v1/blob/f0aae01465e8fcbb36ffa4bbd1f42a89a966a85e079332cea459b6b38df83524
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.
More from nvidia-ai-blueprints/video-search-and-summarization
Every file in nvidia-ai-blueprints/video-search-and-summarization