LLM Torch Profiler Analysis
sgl-project/sglang
Unified LLM torch-profiler triage skill for sglang, vllm, TensorRT-LLM, and TokenSpeed.
A skill your agent uses when picking the searchspace= argument for Search().
$ npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install NVIDIA/CompileIQ compileiq-search-space --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/NVIDIA/CompileIQ.git skills-src && mkdir -p .claude/skills && cp -r skills-src/agent-skills/compileiq-search-space .claude/skills/compileiq-search-space && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "compileiq-search-space" agent skill from https://github.com/NVIDIA/CompileIQ/tree/main/agent-skills/compileiq-search-space into .claude/skills/compileiq-search-space/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "compileiq-search-space", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/NVIDIA/CompileIQ/tree/main/agent-skills/compileiq-search-spaceType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install NVIDIA/CompileIQ compileiq-search-space --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NVIDIA/CompileIQ.git skills-src && mkdir -p .agents/skills && cp -r skills-src/agent-skills/compileiq-search-space .agents/skills/compileiq-search-space && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "compileiq-search-space" agent skill from https://github.com/NVIDIA/CompileIQ/tree/main/agent-skills/compileiq-search-space into .agents/skills/compileiq-search-space/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "compileiq-search-space", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install NVIDIA/CompileIQ compileiq-search-space --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NVIDIA/CompileIQ.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/agent-skills/compileiq-search-space .cursor/skills/compileiq-search-space && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "compileiq-search-space" agent skill from https://github.com/NVIDIA/CompileIQ/tree/main/agent-skills/compileiq-search-space into .cursor/skills/compileiq-search-space/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "compileiq-search-space", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/NVIDIA/CompileIQ.git --path agent-skills/compileiq-search-space--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install NVIDIA/CompileIQ compileiq-search-space --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NVIDIA/CompileIQ.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/agent-skills/compileiq-search-space .gemini/skills/compileiq-search-space && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "compileiq-search-space" agent skill from https://github.com/NVIDIA/CompileIQ/tree/main/agent-skills/compileiq-search-space into .gemini/skills/compileiq-search-space/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "compileiq-search-space", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install NVIDIA/CompileIQ compileiq-search-spaceInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/NVIDIA/CompileIQ.git skills-src && mkdir -p .github/skills && cp -r skills-src/agent-skills/compileiq-search-space .github/skills/compileiq-search-space && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "compileiq-search-space" agent skill from https://github.com/NVIDIA/CompileIQ/tree/main/agent-skills/compileiq-search-space into .github/skills/compileiq-search-space/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "compileiq-search-space", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install NVIDIA/CompileIQ compileiq-search-space --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NVIDIA/CompileIQ.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/agent-skills/compileiq-search-space .opencode/skills/compileiq-search-space && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "compileiq-search-space" agent skill from https://github.com/NVIDIA/CompileIQ/tree/main/agent-skills/compileiq-search-space into .opencode/skills/compileiq-search-space/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "compileiq-search-space", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
compileiq-search-spaceA skill your agent uses when picking the searchspace= argument for Search().
Compileiq Search Space is an agent skill from NVIDIA/CompileIQ, published by the product's own GitHub organization. Use when picking the searchspace= argument for Search(). Covers the three provider classes (PtxasSearchSpace, NvccSearchSpace, LocalSearchSpaceBin), how to pin a version/variant/tag, the attention-focused 'att' variant for attention kernels (FlashAttention, GQA, MHA, MLA, FlashInfer Batch Decode), air-gapped mirroring via CIQSEARCHSPACESDIR, and custom user-defined search spaces built from compileiq.searchspaces.base primitives. Triggers on "search space", "PtxasSearchSpace", "NvccSearchSpace", "air-gapped…
Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It works with NVIDIA AI Platform. The repository describes itself as: An Optimizer for Nvidia Compilers. The licence is Apache-2.0.
Read from SKILL.md and the folder at commit 743aca4. It shows what the files ask for, not the result of running them.
Pre-approves these tools, so the agent can use them without asking each time:
BashReadFrom allowed-tools in the SKILL.md frontmatter.
Shell commands in SKILL.md call:
pythonghFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md. Its commands use gh, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Compileiq Search Space loads about 1.9k tokens when it runs. Until then it costs about 158 tokens; SKILL.md has 485 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check noted patterns worth knowing about, such as sudo or a known installer.
allowed-tools: Bash, ReadAutomated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from NVIDIA/CompileIQ at commit 743aca4, republished under its Apache-2.0 licence (© NVIDIA). 485 words, ~1,932 tokens.
.claude/skills/compileiq-search-space/SKILL.md (or your agent's skills folder).CompileIQ ships compiler search spaces as release-backed binary blobs that
the providers in compileiq.search_spaces.compilers fetch on demand. This
skill covers the three provider classes, the variants they expose today,
how to use them offline, and how to define a custom search space for
non-compiler tuning.
.bin you have on hand.Reference: compileiq/search_spaces/compilers.py:66-102.
from compileiq.search_spaces.compilers import (
PtxasSearchSpace,
NvccSearchSpace,
LocalSearchSpaceBin,
)
# Default — auto-fetch latest PTXAS 13.3 default variant from GitHub releases
ss = PtxasSearchSpace()
# Pinned for reproducibility
ss = PtxasSearchSpace(version="13.3", variant="default", tag="search-spaces-2026.05")
# **Attention** workloads (FlashAttention / GQA / MHA / MLA / FlashInfer Batch Decode)
ss = PtxasSearchSpace(version="13.3", variant="att")
# NVCC variant for full-pipeline tuning
ss = NvccSearchSpace(version="13.3")
# Single .bin you already have on disk — skips manifest + network
ss = LocalSearchSpaceBin("/path/to/ptxas13.3_search_space.bin")
# Pass to Search
from compileiq.ciq import Search
tuner = Search(objective_function=..., search_space=ss, search_config=...)From release/search-spaces/manifest-source.yaml:
| Compiler | Version | Variant | File | When to use |
|---|---|---|---|---|
ptxas | 13.3 | default | ptxas13.3_search_space.bin | Generic PTXAS tuning; the right starting point for most kernels. |
ptxas | 13.3 | att | ptxas13.3_att_search_space.bin | Attention workloads. att is short for attention, not "attribute". This variant is curated for FlashAttention, GQA, MHA, MLA, FlashInfer Batch Decode, and similar attention kernels. Prefer this whenever the kernel is attention-shaped. |
nvcc | 13.3 | default | nvcc13.3_search_space.bin | Full-compiler (front-end + back-end) tuning when you want NVCC-level knobs, not just PTXAS. |
To enumerate variants in the latest release at any time:
gh release view --json assets --jq '.assets[].name' -R NVIDIA/CompileIQ <tag>Pre-download the manifest plus all .bin files on a connected host, then point
CompileIQ at the local mirror via CIQ_SEARCH_SPACES_DIR:
# On a connected host
mkdir -p /shared/ciq-search-spaces
gh release download search-spaces-latest -R NVIDIA/CompileIQ -D /shared/ciq-search-spaces
# Move the directory to the air-gapped host (rsync, scp, sneaker-net, …)
# On the air-gapped host
export CIQ_SEARCH_SPACES_DIR=/shared/ciq-search-spaces
python -c "from compileiq.search_spaces.compilers import PtxasSearchSpace; print(PtxasSearchSpace().retrieve())"Other env-var knobs:
CIQ_SEARCH_SPACES_REPO (default NVIDIA/CompileIQ): override the GitHub
repo the resolver queries — useful for staging or forks.CIQ_SS_TAG_PREFIX (default search-spaces-): tag prefix used when
resolving tag="latest". Rarely needs changing.Resolved binaries are cached at ~/.cache/compileiq/<tag>/<sha256_prefix>_<filename>.
The resolver verifies cached files by sha256; a corrupted entry is re-downloaded
on the next call. Safe to wipe — wiping just costs one re-download.
For hyperparameter tuning, autotuner knobs, or any user-defined space, pass
a dict (or list-of-dicts) directly instead of a provider. Primitives live in
compileiq/search_spaces/base.py:
import compileiq.search_spaces.base as ss
search_space = {
"block_size": ss.choice([64, 128, 256, 512]),
"unroll": ss.range(start=1, end=8, step=1),
"use_shmem": ss.literal(True, knockout_prob=0.5),
"lr": ss.log_sampling(start=1e-5, end=1e-1, total=20),
}| Primitive | Purpose |
|---|---|
choice([...]) | Sample uniformly from a discrete list. |
range(start, end, step) | Range-like sampling (also supports float steps). |
literal(value, knockout_prob=...) | Constant value; knockout_prob lets the GA disable this parameter. |
log_sampling(start, end, total) | Logarithmic distribution between start and end with total discrete buckets. |
Mixed user + compiler search space (list shape):
search_space = [
{"config_idx": ss.range(0, len(CONFIGS) - 1)}, # user-defined
PtxasSearchSpace(version="13.3"), # compiler-side
]When the search space is a list, the objective receives a list of the same
length — see compileiq-author-objective for the unpacking pattern.
# Default variant resolves
python -c "
from compileiq.search_spaces.compilers import PtxasSearchSpace
p = PtxasSearchSpace().retrieve()
assert p.exists() and p.stat().st_size > 0, p
print(f'default OK: {p}')
"
# Attention variant resolves
python -c "
from compileiq.search_spaces.compilers import PtxasSearchSpace
p = PtxasSearchSpace(version='13.3', variant='att').retrieve()
assert p.exists() and p.stat().st_size > 0, p
print(f'att OK: {p}')
"
# Misconfigured air-gap mirror produces a clear error
CIQ_SEARCH_SPACES_DIR=/nonexistent python -c "
from compileiq.search_spaces.compilers import PtxasSearchSpace
try:
PtxasSearchSpace().retrieve()
print('UNEXPECTED: should have failed')
except Exception as e:
print(f'expected failure: {type(e).__name__}')
"att is attention, not "attribute". Earlier docs and a test fixture
mislabeled the variant as "attribute-based register allocation" — corrected
in this same branch. When recommending the variant to a user, say
"attention" explicitly..acf candidates to apply outside a search. Don't
feed a .acf from a Booster Pack to PtxasSearchSpace(...).LocalSearchSpaceBin doesn't validate .bin contents. It only
validates that the file exists. A malformed file will fail downstream when
the core tries to parse it; the resulting error message points at the
resolver, not the provider.compileiq-author-objective.compileiq-run-search.© NVIDIA, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Just SKILL.md in agent-skills/compileiq-search-space of NVIDIA/CompileIQ.
Open the folder on GitHubat commit 743aca4
Compileiq Search Space next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Compileiq Search Space this skillNVIDIA/CompileIQ | 138 | — | ~1.9k | Automated safety check: Notes | Apache-2.0 | |
| LLM Torch Profiler Analysissgl-project/sglang | 37k | 2 repos | ~6.4k | Automated safety check: Pass | Apache-2.0 | |
| Skill InspectorNVIDIA/SkillSpector | 20k | — | ~1.8k | Automated safety check: Pass | Apache-2.0 | |
| Megatron-LM Container and Dependency SetupNVIDIA/Megatron-LM | 18k | — | ~2.6k | Automated safety check: Pass | Apache-2.0 | |
| Embeddings via 9Routerdecolua/9router | 30k | — | ~604 | Automated safety check: Pass | MIT | |
| Megatron-LM Base Image BumpNVIDIA/Megatron-LM | 18k | — | ~2.8k | Automated safety check: Pass | Apache-2.0 |
sgl-project/sglang
Unified LLM torch-profiler triage skill for sglang, vllm, TensorRT-LLM, and TokenSpeed.
NVIDIA/SkillSpector
Decides whether an agent skill is safe to install by combining a SkillSpector static scan with the agent's own source review, ending in APPROVE, CAUTION or REJECT.
NVIDIA/Megatron-LM
Walks an agent through working inside the Megatron-LM CI container and changing dependencies with uv, so lock files resolve the same locally and in CI.
decolua/9router
Generates vector embeddings through the 9Router /v1/embeddings endpoint, using models from providers such as OpenAI, Gemini, Mistral and Voyage for RAG and semantic search.
NVIDIA/Megatron-LM
Moves Megatron-LM CI to a newer NVIDIA PyTorch base image, updating both the GitHub and GitLab pins together and handling the CI follow-up.
NVIDIA/NemoClaw
Remove bracketed NemoClaw tags from GitHub issue and PR titles.
NVIDIA/CompileIQ
Use BEFORE running a full CompileIQ search. An agent skill from NVIDIA/CompileIQ.
NVIDIA/CompileIQ
A skill your agent uses when starting a fresh CompileIQ project, hitting a socket timeout, or before running any other compileiq- skill.
NVIDIA/CompileIQ
A skill your agent uses when something is wrong: Search() hangs, all evaluations return INVALIDSCORE, scores aren't improving, every config returns the same number, ptxas errors fill the log, CV% is…
NVIDIA/CompileIQ
A skill your agent uses when composing the Search(...) call and calling .start().
NVIDIA/CompileIQ
Use AFTER a Search has completed and BEFORE claiming any speedup or shipping an ACF.
NVIDIA/CompileIQ
A skill your agent uses when writing the objectivefunction= passed to Search().
Works with
A skill your agent uses when picking the searchspace= argument for Search(). Compileiq Search Space is an agent skill from NVIDIA/CompileIQ, published by the product's own GitHub organization. Use when picking the searchspace= argument for Search().
Compileiq Search Space fits situations like: picking the searchspace= argument for Search(); ptxasSearchSpace; nvccSearchSpace; air-gapped compileiq.
Run `npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a claude-code`. Or copy the skill folder (agent-skills/compileiq-search-space in NVIDIA/CompileIQ) into .claude/skills/compileiq-search-space in your project. Claude Code loads it when a task matches its description.
Run `npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a codex`. Or copy the skill folder (agent-skills/compileiq-search-space in NVIDIA/CompileIQ) into .agents/skills/compileiq-search-space in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add NVIDIA/CompileIQ --skill compileiq-search-space -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/compileiq-search-space, .gemini/skills/compileiq-search-space, .github/skills/compileiq-search-space and .opencode/skills/compileiq-search-space in your project.
Going by SKILL.md and its folder, Compileiq Search Space needs the command-line tools its instructions call (python and gh). Our summary lists: Python 3. Its frontmatter pre-approves these tools: Bash, Read.
SKILL.md contains no URLs. Its commands use gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.
Compileiq Search Space is published under the Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.
About 1.9k tokens (SKILL.md is roughly 7.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Compileiq Search Space: LLM Torch Profiler Analysis (sgl-project/sglang, 37k stars), Skill Inspector (NVIDIA/SkillSpector, 20k stars), Megatron-LM Container and Dependency Setup (NVIDIA/Megatron-LM, 18k stars) and Embeddings via 9Router (decolua/9router, 30k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
NVIDIA (a GitHub organization, an official publisher) maintains it in NVIDIA/CompileIQ, which has 138 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on September 23, 2026.
Source: NVIDIA/CompileIQ on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.